Programming in Java
A C Norman, Lent Term 2007
Part IA
2
Contents
1 Preface
1.1 What is programming about? . . . . . . . . .
1.2 What about good programming? . . . . . . .
1.3 Ways to save time and effort . . . . . . . . .
1.3.1 Use existing resources . . . . . . . .
1.3.2 Avoid dead-ends . . . . . . . . . . .
1.3.3 Create new re-usable resources . . . .
1.3.4 Documentation and Test trails . . . .
1.3.5 Do not make the same mistake twice .
1.4 Where does Java fit in? . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2 General advice for novices
7
7
8
9
9
10
10
10
10
12
13
3 Introduction
3.1 Introduction . . . . . . . . . . . . . . .
3.1.1 Books . . . . . . . . . . . . . .
3.2 Practical work . . . . . . . . . . . . . .
3.2.1 Exercises . . . . . . . . . . . .
3.3 A Cook-book Kick-start . . . . . . . .
3.3.1 Code Layout . . . . . . . . . .
3.3.2 Emacs . . . . . . . . . . . . . .
3.3.3 Drawing to a window: JApplets
3.3.4 HTML and appletviewer . . . .
3.3.5 Exercises . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
15
15
18
21
24
28
33
34
37
42
43
4 Basic use of Java
4.1 Data types, constants and operations
4.1.1 Reserved Words . . . . . .
4.1.2 Basic Types . . . . . . . . .
4.1.3 Exercises . . . . . . . . . .
4.2 Operators and expressions . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
49
49
49
51
65
71
3
.
.
.
.
.
.
.
.
.
.
CONTENTS
4
4.3
4.4
4.5
4.6
4.7
4.8
5
4.2.1 Exercises . . . . . . . . . . . . . . .
Control structures . . . . . . . . . . . . . . .
4.3.1 Exercises . . . . . . . . . . . . . . .
Control structures Part 2 . . . . . . . . . . .
4.4.1 Expression Statements . . . . . . . .
4.4.2 Blocks . . . . . . . . . . . . . . . .
4.4.3 Null statements . . . . . . . . . . . .
4.4.4 if . . . . . . . . . . . . . . . . . . .
4.4.5 while, continue and break . . .
4.4.6 do . . . . . . . . . . . . . . . . . . .
4.4.7 for . . . . . . . . . . . . . . . . . .
4.4.8 switch, case and default . . .
4.4.9 return . . . . . . . . . . . . . . .
4.4.10 try, catch and throw, finally
4.4.11 assert . . . . . . . . . . . . . . .
4.4.12 Variable declarations . . . . . . . . .
4.4.13 Method definitions . . . . . . . . . .
4.4.14 Exercises . . . . . . . . . . . . . . .
Java classes and packages . . . . . . . . . . .
4.5.1 Exercises . . . . . . . . . . . . . . .
Inheritance . . . . . . . . . . . . . . . . . .
4.6.1 Inheritance and the standard libraries
4.6.2 Name-spaces and classes . . . . . . .
4.6.3 Program development with classes . .
Generics . . . . . . . . . . . . . . . . . . . .
4.7.1 Exercises . . . . . . . . . . . . . . .
Important features of the class libraries . . . .
4.8.1 File input and output . . . . . . . . .
4.8.2 Big integers . . . . . . . . . . . . . .
4.8.3 Collections . . . . . . . . . . . . . .
4.8.4 Simple use of Threads . . . . . . . .
4.8.5 Network access . . . . . . . . . . . .
4.8.6 Menus, scroll bars and dialog boxes .
4.8.7 Exercises . . . . . . . . . . . . . . .
Designing and testing programs in Java
5.1 Different sorts of programming tasks . .
5.2 Analysis and description of the objective
5.2.1 Important Questions . . . . . .
5.2.2 Informal specifications . . . . .
5.2.3 Formal descriptions . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
74
77
77
82
82
82
83
83
84
84
85
85
87
87
88
88
89
90
98
108
115
116
120
125
129
130
139
140
147
150
150
153
155
160
.
.
.
.
.
167
171
179
179
180
181
CONTENTS
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10
5.11
5.12
5.13
5.14
5.15
5.16
5.17
5.18
5
5.2.4 Executable specifications . . . . . . . .
Ethical Considerations . . . . . . . . . . . . .
How much of the work has been done already?
What skills and knowledge are available? . . .
Design of methods to achieve a goal . . . . . .
5.6.1 Top-Down Design . . . . . . . . . . .
5.6.2 Bottom-Up Implementation . . . . . .
5.6.3 Data Centred Programming . . . . . .
5.6.4 Iterative Refinement . . . . . . . . . .
5.6.5 Which of the above is best? . . . . . .
How do we know it will work? . . . . . . . . .
While you are writing the program . . . . . . .
Documenting a program or project . . . . . . .
How do we know it does work? . . . . . . . . .
Is it efficient? . . . . . . . . . . . . . . . . . .
Identifying errors . . . . . . . . . . . . . . . .
Corrections and other changes . . . . . . . . .
Portability of software . . . . . . . . . . . . .
Team-work . . . . . . . . . . . . . . . . . . .
Lessons learned . . . . . . . . . . . . . . . . .
Final Words . . . . . . . . . . . . . . . . . . .
Challenging exercises . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
181
182
183
185
186
186
189
190
190
191
191
194
195
197
200
201
204
205
206
207
208
208
6 A representative application
219
6.1 A Lisp interpreter . . . . . . . . . . . . . . . . . . . . . . . . . . 219
6.1.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . 233
7 What you do NOT know yet
8 Model Examination Questions
8.1 Java vs ML . . . . . . . .
8.2 Matrix Class . . . . . . . .
8.3 Hash Tables . . . . . . . .
8.4 Compass Rose . . . . . . .
8.5 Language Words . . . . .
8.6 Exception abuse . . . . . .
8.7 Queues . . . . . . . . . .
8.8 Loops . . . . . . . . . . .
8.9 Snap . . . . . . . . . . . .
8.10 Partitions . . . . . . . . .
8.11 Laziness . . . . . . . . . .
235
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
237
237
238
238
239
239
240
240
240
240
241
241
CONTENTS
6
8.12
8.13
8.14
8.15
8.16
8.17
8.18
8.19
8.20
8.21
8.22
8.23
8.24
8.25
8.26
8.27
9
Cryptarithmetic . . . . . . .
Bandits . . . . . . . . . . .
Exception . . . . . . . . . .
Features . . . . . . . . . . .
More features . . . . . . . .
Debate . . . . . . . . . . . .
Design . . . . . . . . . . . .
Filter (Coffee?) . . . . . . .
Parse trees . . . . . . . . . .
Big Addition . . . . . . . .
Lists in Java . . . . . . . . .
Pound, Shillings and Ounces
Details . . . . . . . . . . . .
Name visibility . . . . . . .
Several Small Tasks . . . . .
Some Tiny Questions . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Java 1.5 or 5.0 versus previous versions
9.1 An enhanced for loop . . . . . . . . . . .
9.2 Generics . . . . . . . . . . . . . . . . . . .
9.3 assert . . . . . . . . . . . . . . . . . . .
9.4 Static imports . . . . . . . . . . . . . . . .
9.5 Auto-boxing . . . . . . . . . . . . . . . . .
9.6 Enumerations . . . . . . . . . . . . . . . .
9.7 printf . . . . . . . . . . . . . . . . . . .
9.8 Scanner . . . . . . . . . . . . . . . . . . .
9.9 Variable numbers of arguments for methods
9.10 Annotations . . . . . . . . . . . . . . . . .
9.11 Enhanced concurrency control . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
242
242
244
245
245
246
246
246
247
248
248
248
249
250
250
251
.
.
.
.
.
.
.
.
.
.
.
253
253
254
254
254
254
254
255
255
255
256
256
Chapter 1
Preface
1.1 What is programming about?
There are two stories you can tell yourself about what this course is going to do for
you. The first is the traditional one that it is so you can learn some Java. Acquire
knowledge and skills. The second, which may be more interesting, is to see this
course as part of your journey as you start to become (or at least appreciate what
it is to be) a Computer Scientist. This second perspective suggests that there may
be something for you here whether or not you believe you are already skilled in
Java, and it challenges you to look beyond the mere details to the tought patterns
that link them together.
In the early days of computers programming involved a full understanding of
the way that the hardware of your computer worked, your program, when run,
took over essentially the whole machine and it had to include everything needed
to manage input and output. In extreme cases one started the process of loading code into a computer by using hand-switches to place bit-patterns directly
into the machine’s memory. After a while operating systems came along and
provided serious insulation from that level of extreme awareness of hardware, and
high-level languages make it possible to express programs in at least semi-humanunderstandable form. But still the emphasis was on “writing a program”, which
tended to be a stand-alone application that solved some problem.
Libraries of pre-written sub-programs grew up, but for a very long time the
ones that anybody could rely on having access to were either rather specialist or
the functionality that they provided was at a rather low and boring level. There
were libraries that could really help you with serious tasks (such as building a
windowed user-interface) but none of them gained really global acceptance, and
only a few were of any use on more than one brand of computer. The libraries that
were standard with typical programming languages provided for fairly limited file
7
8
CHAPTER 1. PREFACE
and terminal access input and output, modest string handling and really not a lot
else. Operating systems made their capabilities available in the form of libraries
that programs could call on, but overall coherent design was rare and use of these
“libraries” led to inherently non-portable code.
Building a new library was not part of the common experience of programmers, and indeed large-scale re-use of code was the exception rather than the rule.
There has been an ideal or a dream of re-usable software components for ages,
but it is only recently that it has started to become something that can be not
just feasible but reasonably convenient. Java is one of the languages that encourages this move, and the whole Object Oriented Programming movement that Java
forms part of provides a context.
So in the old world one thought of a program as a large complicated thing
that called upon facilities from a few fixed libraries that you happened to have
available. Today instead of that you should often start a project with the intention
of developing a set of new re-usable and general libraries that themselves build on
and extend existing software components. You will design these libraries so that
once they exist the program you had to write becomes a fairly simple application
of them: it will do some minor customisation and link together different units
within the overall structure of your libraries, but with luck it will of itself be
fairly small and straightforward. If you do this well you will find that the library
you have created will serve you well in future projects, or it may even become
something worth circulating (or selling) of itself. With these ideas in mind you
will want to make it well-structured, robust and you may even feel motivated to
accompany it with some coherent documentation!
So overall the mind-set for the 21st Century is that you design and write reusable components and libraries, and that writing mere stand-alone programs is a
terribly old-fashioned and dull thing to do!
1.2 What about good programming?
The first and utterly overriding character of a good program is that it must be fit
for its purpose. Good programming must not only lead to a good program, but
should do so in a way that reaches a successful conclusion reliably and without
taking more time and effort than is really required.
These comments may seem bland and self-evident, but they have real consequences! The first is that you can not judge a program until you know what its
purpose is. Even though almost all the exercises you will do this year will be
both small and will never have any part of their code re-used it will be proper for
you to practise writing them as if they are much larger and more important. That
will mean that you are expected to accompany the code you write with both notes
1.3. WAYS TO SAVE TIME AND EFFORT
9
about its external behaviour and how to use it and with comments that describe its
internal structure and organisation. For certain sorts of library code it will make
sense to use the documentation arrangements that the main Java libraries use. This
involves things called “documentation comments” and a utility called javadoc
that will be described later.
Without this documentation you may believe that your programs meet their
purpose but you do not have any basis for expecting others the agree.
1.3 Ways to save time and effort
Working with computers can swallow up an astonishing amount of time. To be
able to get everything you need done you will want to find ways of economising.
The key to doing this effectively is to concentrate on techniques that save time in
the long run. Some ideas that appear to speed things up in the short run can end
up costing more later on!
1.3.1 Use existing resources
You are encouraged to use code-fragments from these notes in any way you want.
You can sometimes get a fresh project off the ground by extracting at least fragments from a previous piece of work you have done. The Java libraries are your
friend: they contain facilities to do many of the things you will find yourself needing. In general before you ever write anything from scratch for yourself consider
whether there is something that can give you a head-start.
Everybody might reasonably worry that the above paragraph could be seen as
an invitation to plagiarise! Do not take it that way: couple it with a very firm remark that when you use other material you should acknowledge your sources, and
you should not pillage the material of those who are unwilling to make their work
available to you. As far as tickable exercises for this course are concerned you are
encouraged to discuss what you are doing with friends and supervisors, and collect code-sketches and fragments from them, provided that when you submit your
work to the department you really understand everything in your submission and
you have learned enough that (if necessary) you could then instantly and comfortable re-create your submission in a sound-proof booth visibly cut-off from further
help. So rather than sit and suffer in isolation, seek web-sites, friends, demonstrators, books and code libraries to give you guidance so long as you learn from then
and do not just blindly copy!
10
CHAPTER 1. PREFACE
1.3.2 Avoid dead-ends
Sometimes you can start designing or writing some code and as you go things
seem to get harder and harder. Something is not working and you have no idea
why. You do not want to get in such a state! Clear advance planning and a well
organised set of working habits are the best way to avoid the mess. If you find
yourself in what feels like a dead-end then avoid (a) panic (b) a tendency to try
almost random changes in the hope that things will improve and (c) temptation
to work all afternoon, evening and night until you solve things. Go back and
look at your plan. If necessary refine it so you can make progress in tiny steps.
Explain your plan and your code to somebody else (either in person or in the form
of written documentation). But do not just get bogged down: taking a break and
coming back fresh can often save overall time.
1.3.3 Create new re-usable resources
An ideal that this course would like to instil in you is one of creating re-usable
bodies of code. This will take more care and time when you first implement them
(and of course anything re-usable deserves proper documentation and testing) but
that can be well paid back when you get a chance to call on it again. At a minimum this can include keeping all your working code from both the tickable and
other exercises in these notes so you can use parts of them as templates in future
projects.
1.3.4 Documentation and Test trails
Neatly formatted code with clear comments and a well set out collection of test
cases can seem slower to write then a jumble of code that is just thrown together.
However long experience suggests that the jumble of code is much less likely to
work first time, and that especially as your projects get bigger that early investment
in good habits pay dividends.
1.3.5 Do not make the same mistake twice
Especially while learning a new language, such as Java, you will make mistakes.
As you design and write gradually larger and larger bodies of code you will make
mistakes. Observe and appreciate these, and try to observe yourself as you uncover and correct them. Possibly even keep a small notebook of “bugs I have had
in my code”. Then each time you make a mistake seek some scheme that can prevent the same one from causing significant trouble in the future. Your fingers will
always leave typos in everything you write and your mind can always wander: the
1.3. WAYS TO SAVE TIME AND EFFORT
11
idea is not to avoid glitches totally, it is to build up a personal toolkit of ways to
overcome them without pain or waste.
CHAPTER 1. PREFACE
12
1.4 Where does Java fit in?
There are those who believe that Object Oriented Design and Programming is The Answer to reliable
large-scale system building, a silver bullet1 that cures
the major woes of the last fifty years of over-costly and
haphazard use of computers. Java is one of the major
practical and widely-used languages that fall within
the Object Oriented family. Key attitudes that come
with this are that projects should be structured into potentially re-usable blocks (the Java class construct
that you will learn about later being a major way of
achieving this). These blocks should each take responsibility for just one aspect of the overall behaviour you
are trying to code up. The decomposition should be
arranged so that interaction between blocks is as tidy
and disciplined as possible.
Overall at least a rough caricature is that ML Figure 1.1: Silver Bullet
stresses absolute correctness via mathematically styled Needed.
structure, and encourages very concise programming
styles. Java on the other hand follows a view that language constructs that support
large-scale structuring of projects are the key. It also expects that having the user
write out types and qualifiers explicitly will help others to read your program.
ML as taught last term provides a fairly basic library, but mostly you spend the
Michaelmas Term writing stand-alone programs and fragments. With Java there
is heavy emphasis on a rich (and perhaps hence complicated) library that supports
a very full range of computing needs.
1 Brad
words.
Cox in Byte magazine October 1990, pp 209–218 puts things in much these extreme
Chapter 2
General advice for novices
Following tradition, I provide ten items of guidance for the benefit of those who
are relatively new to programming. I hope that each of these will be re-inforced
during the course as a whole, but here they are collected together at the beginning:
1 Understand the task you are about to solve before starting to write a program about it. Work through methods and procedures by hand on paper
etc. Plan some test cases. Identify cases that will represent boundaries
or oddities. In general prepare a plan before you start going anywhere
near a computer;
2 Sketch the structure of the whole of your code out informally so you have
full overview before fussing about exact syntax etc. Ensure you know
what you expect that the computer will do. This initial sketch can be
very informal, and may be in terms of diagrams rather than anything that
looks much like real programming. The key word here is “structure”.
This applies with way greater force when your code starts to grow: you
should always design a good way to factor your code into reasonably
self-contained and independent components (each will be one “class” in
your code) right from the start;
13
14
CHAPTER 2. GENERAL ADVICE FOR NOVICES
3 Write out key parts of above in the form of comments before you start
the real code. Concentrate in these comments on the “what” and “why”
of your code rather the details of “how”. This will really help when you
show your work to somebody else because you need help! I will explain
this one again: The first thing you will type into a computer when you
start writing any program will be a set of overview comments that explain
its strategy and structure;
4 At least for a first version of anything, favour clarity and obvious correctness over pretty well everything else. Clever tricks, worries about
efficiency, generalisations etc can come later;
5 Neat consistent layout and thoughtfully named fields, methods, variables
etc. are a good investment of your time. Cryptic is bad even if it saves
keystrokes in the short term;
6 If a task is too big to solve in just one gulp look for ways of breaking it
down into sub-tasks. As you do this think about ways you will be able to
test code you write for each sub-task and work on the whole thing step
by step;
7 When you try to compile your code and see a syntax error do not panic.
Learn to interpret the compiler’s diagnostics. And only try to remove
one error at a time: count it as a success if next time you try to compile
the first error has give so you can then concentrate on the second;
8 When you have compiled your program and run it and it gives wrong
answers or behaves badly do not panic. First work to understand what is
wrong and only after you have found where the problem is think about
ways to fit it. Do not just try random changes! Eg. confirm what your
program actually does by adding assert and extra print statements;
9 Whenever you find you have to change your program review comments,
consider if it will now do exactly what you want, and re-run all your test
cases. Experience shows that changes (for whatever cause) can introduce
new problems while you are in the process of fixing old ones;
10 If you find you are spending a seriously long time trying to make sense
of anything then find help from friends or a supervisor or a book. Do not
just keep building up your frustration not getting anywhere!
Chapter 3
Introduction
3.1 Introduction
We have been using Java as a first-year teaching language here in Cambridge
since 1997-8. We teach this course following on from “Foundations of Computer
Science” which used ML, and there are a number of things it is intended to do:
1. Provide all of our students with exposure to a common programming language that can be used by later courses and practical work in the CST;
2. Introduce the syntax that is (almost) common to several of the most widely
used practical programming languages today (the syntax of Java has a great
deal in common with that of C and C++, so having learned Java you are
quite a long way to understanding those languages too);
3. Discuss the process of designing, writing and debugging programs and raise
some awareness of issues of style;
4. Present the Object Oriented aspects of a programming language as means
to enforce modularity in large programs;
5. Teach basic use of Java, a language that has significant relevance in the
outside world today.
Note that in our Part IA course “Software Engineering II” provides significant extra coverage on issues of structuring programs (especially ones that are large or
developed by collaborative work in a group), and in Part IB course there is a lecture course once entitled “Further Java” and now renamed ”Concurrent Systems
and Applications”: it should not be imagined that I will cover all aspects of the
language or its use here!
15
16
CHAPTER 3. INTRODUCTION
The nature of teaching a course involving programming in some particular language means that some features need to be mentioned well before the place where
they can be fully explained, and so it will not make sense to keep the presentation
in lectures totally linear and tied to these notes, but for supervision purposes the
structure shown here should suffice. With each section I will have a few examples
or exercises. Especially at the start of the course these will often be pretty silly,
but the ones right at the end can be viewed as samples of the sort of question that
might arise in the examination. Although I want some of my examples to be nice
and easy I would like to have others that are interesting challenges for those who
already think they know it all (ha ha). It is always very hard to judge the amount
of trouble these will give you all, so if they are either too easy or too difficult
I apologise. Examination questions will be set on the supposition that you have
attempted a reasonable sampling of the exercises.
The aim of these notes is that they should serve both as guidance to students
and to supervisors, and so there is no separate supervisor’s guide. Originally I had
intended that they would be structured into sixteen sections corresponding to the
sixteen lectures available. As I prepared the notes I concluded that such a rigid
arrangement was not tenable. Thus the lectures can be expected to cover roughly
the material in these notes in roughly the same order, with an approximation to
one-sixteenth of the entire notes corresponding to each lecture!
It might be noted that a Java course for the Diploma students runs during the
Michaelmas term. The lecture notes associated with that course may provide a
presentation of Java which is different from mine and thus may complement my
lectures or shed light on issues that I fail to.
The course this year will be based on use of the version of Java sometimes
known as “Java 5.0” and sometimes as “Java 1.5”. This should now be counted as
the current and widely-used version, but if you use computers other than the main
university ones here you may come across earlier releases. Please avoid them for
course-related work to avoid confusion.
Some members of the audience for this course will already have significant
practical experience with Java. Others will have written lots of programs before
but in C, C++ or Pascal, but the only thing I can properly assume here is that
everybody has attended the Foundations of Computer Science course given in the
Michaelmas term and hence that everybody is used to writing code in the language
ML. While those who have seen Java before will undoubtedly find the first few
lectures and exercises here very easy, I hope that they will find material that is new
and worth-while being introduced in due course. In the first year that this course
was given it was observed by one Director of Studies at a large College that some
of his students who did already know Java concluded on that basis that they need
not attend the lectures, but that their examination results indicated that this had
not been a perfect judgement call.
3.1. INTRODUCTION
Figure 3.1: Reproduced courtesy Kevin McCurley.
17
18
CHAPTER 3. INTRODUCTION
3.1.1 Books
All bookshops these days seem to devote many metres of shelf-space to books that
purport to teach you Java in a given small number of days, to help you even if you
are an “idiot”, or to provide “comprehensive and detailed” coverage of even those
parts of Java that the language definers have left deliberately vague. I believe that
this is a course where it is important for every student to have their own copy
of a supporting textbook/manual. But the issue of which book to buy will end
up a somewhat personal choice since differing levels of detail will suit different
students! Browse the following in libraries and bookshops, talk to students in
higher years and seek advice from your Directors of Studies and Supervisors about
what is liable to suit you.
My first recommendation has as most of its pages what is in effect hard copy
of the on-line detailed documentation of the Java library. As a result it is not a
smooth consistent read, but I find that very many people need that information in
printed form while they are getting used to navigating the library and understanding what it can do for them.
Java in a Nutshell fifth edition (Feb 2005)
Java Foundation Classes in a Nutshell (1999)
David Flanagan
O’Reilly
Note that the fifth edition is due to be published as your course begins!
There are two books[11, 6] that I think you might reasonably consider and that
are probably easier for self study in that they do not get so rapidly enmeshed in
full detail.
Thinking in Java
Bruce Eckel
Prentice-Hall, 2002 third edition
and
Java Gently
Judy Bishop
Addison Wesley, 2003, third edition
Eckel’s book is distributed (at no cost) via www.eckelobjects.com so if
you are actually prefer reading computer screens to real books it counts as a great
bargain!
Some Directors of Studies will strongly point you towards the book[2] that
was that main text for this course a couple of years ago:
Objects First with Java: a Practical Introduction using BLUEJ
David Barnes and Michael Kölling
Prentice Hall/Pearson, 2005, second edition
3.1. INTRODUCTION
Figure 3.2: Not (quite) the main course book.
19
20
CHAPTER 3. INTRODUCTION
This book emphasises issues of overall program structure and design above
concern for the exact details of the Java language or its libraries and so is almost
exactly the antithesis of the Nutshell books! It was used as the main teaching text
here in 2003-4, and you may find the BlueJ software (http://www.bluej.org)
provides a useful environment within which to develop and test your code. Note
that this year’s edition of both the book and the software has developed from the
versions available last year.
There will be plenty of other useful books, and any individual student may
find a different one especially to their own taste. If selecting a book other than
the one I suggest that you make sure that what you learn from it can be related to
the lectures that I give. Since this year we are using a rather new version of Java
beware that old editions of books may not be sufficiently up to date.
Java is a “buzzword-compliant” language, and when people hear that you are
learning it they will instantly pick up all sorts of expectations. Even though this
course is sixteen lectures long I will not be able to fulfil all of these, and that is in
part why the Computer Science Tripos has a course entitled “Concurrent Systems
and Applications” in Part IB that follows on from this one. There are three issues
that I should mention right here at the start of the notes, if only to protect myself
and the department against misunderstandings as to our purpose:
Java is for use animating Web pages:
Some of the huge first flush of enthusiasm that greeted the emergence of
Java was because it could be used to make rather naff animated figures dance
on web pages. This was of course amazing when web pages had previously
been so rigidly static, but it is not a good model for the central issues in
Computer Science. This will typically not be the sort of use of Java that we
try to teach you here;
Java is the best programming language:
The Computer Laboratory shows by its actions that it views ML as its preference for a first language to teach its students, with Java as a second one.
Later on in the course we will provide coverage ranging from brief mention to detailed explanations of quite a few other languages: certainly C,
C++, Lisp and Prolog. The Software engineering courses mention a scheme
called just ‘Z’ that is in effect a programming language, and you will see
from past examination papers that we have high regard for Modula 3. What
is shown by that is that the Computer Laboratory view is that different languages may prove best for different tasks, and that the optimal choice will
change as the years go by (it happens that we no longer teach our students
either Fortran or COBOL, and our coverage of assembly code is present
3.2. PRACTICAL WORK
21
because it forms an important link between the concerns of hardware designers, operating system experts and compiler writers, and not because we
expect students to do project work in it). At present Java is our choice for
the first “traditional”-style programming language we teach: this does not
mean it will automatically be the only or best choice for all future practical
work and projects;
Students should be taught about “programming in the large”:
As this is a first year course I will be concentrating on the fundamental
building blocks of program construction. This is in line with the Engineering Council “EA1” concern about introducing students to the fundamental
tools, materials and techniques in their subject. I view it is self-evident
that until a student can write small programs competently and painlessly it
would not make sense to expect them to be able to work in groups on large
projects. However in all the practical work associated with this course you
should expect the assessors to demand that all code you write is well laid
out, properly commented, that it displays a sensible programming style and
that you are in a position to justify its correctness. In short that a generally
professional approach has been taken even though many of the exercises are
short and somewhat jokey toy problems.
3.2 Practical work
The main environment the laboratory expects you to use for this course is PWF
Linux. At the start of Term you should be given an introduction that explains how
to re-boot certainly the PWF systems in Cockroft 4 or in the Intel Laboratory in
the Gates Building so they run Linux. PWF workstations in other parts of the
University may not have been configured with this dual-boot option, but if they
have then you can use them. Although Java runs perfectly happily on Windows
we want you to do much of your practical work on Linux so that by the time you
come to the Operating System course later in the year you have made significant
personal use of both Windows and Linux.
At least for the first half of the Term we would also encourage you to use the
emacs editor and build and run your Java programs using the somewhat primitive
command-line driven tools javac, java and appletviewer. Use of these will
be explained later. The reasoning behind this is not that it guarantees to make your
Java-specific experience as comfortable as possible, but because the technologies
involved are ones you need to find out about at some stage! Specifically I note
that
• emacs is a rich and powerful editor. You can use it in a simple way while
CHAPTER 3. INTRODUCTION
22
you are beginning work, but it has extension mechanisms that allow it to
morph to provide specialist support for different sorts of document, and it
can provide a single environment (and set of keystrokes to learn) that covers
not just editing your program but also compiling and running it, reading and
sending e-mail and many other tasks. It probably counts as the most widely
used general-purpose Unix/Linux editor and versions for Windows are also
available. Your really simple use of it now will help those of you who
choose to use if in more elaborate ways later on.
• The use of the javac and java commands explicitly (as distinct from you
using them implicitly through an all-encompassing specialist Java development environment) means that when you see any curious messages or
complaints you know where they come from. It also introduces you to a
typical model for how software is built (the edit, compile, test cycle). When
you are more experienced you will no doubt move on and use integrated
environments1. In some respects these help by doing things for you – but
especially since you have survived the Foundations of Computer Science
course last Term it now seems proper that you get to see how to do things
for yourself.
For reference material it may prove most convenient to use on-line documentation, and in particular the web-browsable HTML version. This is available to you
in $CLTEACH/acn1/java/docs, so you can launch a browser and start looking
at it by going
firefox $CLTEACH/acn1/java/docs/index.html &
and around the first thing you may want to do is to set yourself a bookmark on that
page. There is a huge amount of documentation there. The bits I find most useful
are the “Java 2 Platform API Specification” which documents (in painful detail)
all of the library facilities that are provided, and the “Java Tutorial” which links
to a Sun website with much helpful explanation, and which you may find a very
good complement to the textbooks I have suggested. All the time I am writing
any Java code at all I will have a web-browser open on the “API” section of the
documentation, since it is useful to have a quick way to check details of the library
very close at hand.
You can obviously run PWF Linux in one of the big shared workstation areas,
and there is a great deal to be said for at least starting off that way: you can compare notes with other students when you have problems. But you can also access
1 Microsoft’s
Visual Studio is perhaps a definitive example: for Java you can install either
Netbeans (from Sun) or Eclipse (from IBM) free of charge. BlueJ has very different objectives but
may also prove useful to some.
3.2. PRACTICAL WORK
Figure 3.3: Remember about RSI, posture etc, please.
23
CHAPTER 3. INTRODUCTION
24
PWF by using ssh and an “X-windows server” to access of of the lab’s PWF
linux systems that are set up for remote use, eg linux2.pwf.cl.cam.ac.uk or
linux.pwf.cam.ac.uk. If your own computer is set up to run Linux those will
already be present for you. If you run Windows you can get good versions free of
charge by installing a Unix-compatibility layer from http://www.cygwin.com,
but getting everything to work nicely there may be messy enough that those of a
nervous disposition would do better to work in Cockroft 4 or one of the College
computer rooms where PWF Linux is directly available!
It is also perfectly in order for you to install Java on your own computer. Apart
from the fact that the Java development kit uses around 450 Mbytes installing it
should not prove hard, it does not cost anything and performance should work
well under either Windows, Linux or MacOS on any even reasonably recent pc.
If you do that you must be willing to take full responsibility for installing and
maintaining everything, and should take care to back up all important files. For
just running small Java exercises there should not be much difference in the experience you have using your own rather than a public machine2, however if you
habitually use a PWF system somewhere other than in Cockroft 4 or the Intel
laboratory your own system might reduce your need to wait while you re-boot a
public machine into Linux, and if you experiment with one of the integrated Java
environments you nay find performance much better on your own system. If you
have a Macintosh note that Java 1.5 has only very recently become available, so
please double-check that that is the version you have.
To fetch a Java compiler you will need to connect to
http://java.sun.com/j2se
where you can find the Java “SDK Standard Edition, version 5.0”, and its accompanying documentation. You should be aware that the package you have to
download is around 50 Mbytes for the main kit, with the documentation being an
additional large download and the “Netbeans” development environment yet more
that you may want to explore but are not obliged to worry about. Sun can supply
either Windows (2000/XP) or Linux versions of all of these.
The Eclipse development environment can be found at http://www.eclipse.org.
3.2.1 Exercises
Tickable Exercise 1
The first part of this tickable exercise is issued as part of the introduction to the
use of Linux on the PWF. The task set here is thus “Part B” of the complete Tick.
2 Great
thanks are due to the Computing Service for ensuring that this is the case.
3.2. PRACTICAL WORK
25
Log on to the PWF. Create a new directory and select it as your current one,
eg
mkdir Tick1B
cd Tick1B
Issue the following commands that copy two files form the Computer Lab’s
teaching filespace into your new directory.
cp $CLTEACH/acn1/TickBase.class .
cp $CLTEACH/acn1/TickBase1.class .
You should be able to check that the files are present. These two files provide
a basis upon which the exercise builds.
Now inspect Figure 3.4 which is documentation associated with the two files
that you have just copied. A more extensive version of the same material is available on-line as
www.cl.cam.ac.uk/Teaching/current/ProgJava/notes/TickDoc/
Now prepare a file that called Tick1.java containing the text
// Tick 1.
Your Name Goes Here
public class Tick1 extends TickBase
{
public static void main(String []args)
{
(new Tick1()).setVisible(true);
}
public String myName()
{
return "Your Name";
}
}
Obviously you will put your own name in the places that are suggested by
what I have written here!
Compile your program and then run it:
javac Tick1.java
java Tick1
26
CHAPTER 3. INTRODUCTION
Figure 3.4: Documentation of Tick 1 Part B.
3.2. PRACTICAL WORK
27
If all has gone well a window should appear, and it should have some text
and a pattern on it. There is a menu that you can select. If you copy the files
to your own machine you can try the print menu, but on the PWF there are
technical reasons why that is not supported, and these lie outside just Java. So
select the menu item labelled postscript. You should then see a dialog box
asking you to choose a file name. I suggest that you select the name tick1.ps
and I very strongly suggest that you use the extension .ps whatever name you
actually choose. When you accept the file-name you have chosen the “select file”
dialog box disappears and you can not see that anything much has happened, but
the file you indicated should have been created for you. It should contain an
image of the screen window in the Postscript document format. Close the little
Java window, and you can send this to a printer using the command
lpr tick1.ps
The resulting sheet of paper is what goes to your ticker.
As an optional extra you can arrange to change the colour of (some of) the text
generated by adding lines roughly like the following to your Java source file:
public java.awt.Color myColour()
{
//
RED GREEN BLUE
return new java.awt.Color(0.7f, 0.1f, 1.0f);
}
where the three floating point numbers given (note that you have to write a letter
‘f’ at their end) should each be in the range 0.0 to 1.0 and they give the proportions
or red, green and blue in the colour.
You can also check what happens if you present your name in different ways.
For instance I tried “A C Norman” as well as “Arthur Norman”. If you wanted
to keep your program in a file called say MyTick.java rather than Tick1.java
you would have to change its name within the file too. Verify that you can do that.
Discussion
This exercise is intended to send several signals and messages about Java:
• One can build new programs building on existing components that do quite
a lot for you. Here you copied in the TickBase class files, but your own program then builds on them and can customise the behaviour of the provided
code in various ways. Through doing this a very short fragment of code let
you create a window and print its contents;
CHAPTER 3. INTRODUCTION
28
• To use the software component TickBase you do not need to see its internal
structure: all you need is documentation about how to use it. As part of
stressing this I am not going to provide you with the source code of TickBase, but by the end of this course you will probably be able to re-create
it;
• Part of the way of using components like this involves the Java keyword
extends, and part of the way that the code runs involves the keyword new.
These are both key parts of the Object Oriented structure of Java, and you
should look forward to finding out more about just what they really mean
later on.
• The page of documentation included as part of these notes tells you that
TickBase is interested in a myName(). This documentation is in the style
of the bulk of the Java on-line documentation, and was created by using a
simple tool called javadoc that interpreted some special comments in the
TickBase source code. However the full output from javadoc is on the web
page listed a little earlier and perhaps gives a bit more of an idea of just how
much complexity is involved under the surface. The lesson that I learn is
that if you use javadoc for anything other than a full-scale project you will
need to edit its output heavily to remove material that your audience does
not really need to see.
(End of tickable exercise)
3.3 A Cook-book Kick-start
In this section I will try to get you started with Java. This means that all sorts of
aspects of it will be described in an order that is not really logical, but is motivated
by that fact that some features of the language must be described early if you are
to get any programs at all written. I will not provide much justification for the
recipes that I give. Later on it will be possible to re-visit these examples and
understand what the various odd keywords are all saying and what options might
be available, but for now you can just copy them out parrot fashion.
I would like you to type in all the examples for yourselves and try them out,
since that will educate your fingers into following the rules that Java imposes, and
it will also (each time your fingers stray) give you exposure to Java error messages
and the joys of finding and fixing mistakes.
My first example in fact is an echo of the first part of Java Tick 1. A mildly silly
tradition in teaching programming languages is that the first program presented
should just print out “hello”. The way of doing this in Java looks like this:
3.3. A COOK-BOOK KICK-START
29
System.out.printf("Hello");
which is a call to a library function called printf3 that will display the given
string. The prefix “System.out” is not part of the name of the function — it
happens to be providing the instruction that that printed output should be sent to
the standard output stream, ie typically straight to your screen or terminal. In
essential terms the line shown above is the whole important part of your first Java
program. However there is actually quite a lot more to be discussed before you
can try it!
The first thing is that Java is a compiled programming language, so unlike
the situation you have seen in ML it is essential to place your program in a file
before it can be used. In this case you should use a file called Hello.java and
it is essential that the file name should start with the word Hello since that is the
name that we will soon repeat within the file. The spelling should be with a capital
letter as shown4 , and the file-name should be completed with the suffix .java.
If you start emacs and use the menu selection “Files/Open File” you get a
chance to create a new file for this project, and if you may5 notice that when you
type in the string in the example it is displayed in an alternate colour (to help
remind you to match your quote marks), and when you type the close parenthesis
after the string the matching bracket gets flashed to help you keep that side of
things under control.
It is possible to make very extensive customisations of emacs. If you put a
file called .emacs in your home directory it can contain directives that apply
whenever emacs starts. In particular if you put a line
(global-font-lock-mode t)
then you will get syntax colouring enabled every time: I find this convenient. For
now I suggest that you avoid putting large amounts of other clever stuff there!
You will also see that the menu-bar at the top of the emacs window has entries that let you do all the things that editors ought to — and more besides. See
figures 3.5 and 3.7: note that the printed form of my notes will be in black and
white but the downloadable version on the lab’s web page
http://www.cl.cam.ac.uk/Teaching/2005/ProgJava
will show relevant information in colour. Also note that the sample programs
being edited and tested in the pictures of emacs in use may be ones taken from
previous years’ versions of this course.
3 Many
Java texts use a function println here rather than printf.
actually if you are working on a Windows system the capitalisation is not so important,
but even there you are strongly advised to keep to it so that when you transfer your programs back
to Unix before showing them to the assessors they still work!
5 Provided the “global font lock” options is selected.
4 Well
CHAPTER 3. INTRODUCTION
30
A “Java mode” is automatically selected when you edit a file whose name ends
in .java and this is the first pay-off you see from this convention. If you select
a “global font lock” this can colour your code so that language keywords, strings,
comments and so on are all displayed in different colours6. It also assists with
indentation and provides editing commands that move around in the file in a way
that understands Java syntax. A major feature of emacs is that it is amazingly
customisable, and configuration files can provide it with special support for many
languages and layout conventions. If you browse enough sites on the web you
may find many extra options that you can install: hold back and avoid these until
you have got really used to the default setup! Please!
Figure 3.5: Two windows, with emacs editing a program.
Your complete first Java program needs a great pile of guff that surrounds the
one interesting line we have and turns it into something that can be executed. In
essence two things need to be documented. The first is something that indicates
the external name that the program will be known by. This will always be exactly
the same as the start of the name of the file it is stored in. You may consider it
silly to have to re-state information that looks as if it should already be available,
but for now please suspend disbelief and accept that a program that lives in a file
called Hello.java will have to contain the text
6 At
a minimum this can be very helpful if you accidentally fail to close a string or comment!
3.3. A COOK-BOOK KICK-START
31
public class Hello
{
...
}
where the ... will be filled in soon with material that includes our call to
System.out.println.
The second piece of information needed is an indication of where Java should
start its processing, and the convention that the language imposes here is that it
expects to find a procedure7 with the name main. The definition of a suitable
procedure then involves the incantation
public static void main(String[] args)
{
...
}
of which the only word that is currently worth describing is “main”, which is a
reminder of the historical tendency to refer to the place where a program started as
being the “main program” while what are now known as functions or procedures
might have been called “sub-programs”.
Comments can be introduced by “//” and every good program starts with a
note that explains a few key facts about it. Obviously the longer the program
the more that it will be proper to put in comments at both the start and throughout
your code, but note that assessors will certainly expect your name and the exercise
identification to be at the head of every submission you make.
Putting this all together we get the full contents of the file Hello.java as
// This is the file "Hello.java" prepared by A C Norman
// and the program just prints a fixed message. 1998-2006.
public class Hello
{
public static void main(String[] args)
{
System.out.printf("Hello%n");
}
}
7 In
these notes I will use the terms “function”, “procedure” and “method” pretty-well interchangeably. Some other languages use these words to indicate refined differences — typically
the term “procedure” would be something that did not return a value, while a “function” would.
The word “method” comes out of ideas of so-called Object Oriented Programming and indicated
a function that is defined within a “class”. Although I have not yet explained what a class is we
have seen the keyword class towards the head of our Java programs.
32
CHAPTER 3. INTRODUCTION
Figure 3.6: Style does matter.
3.3. A COOK-BOOK KICK-START
33
There is a yet further odd addition in what I have just shown. The %n arranges
to put a newline at the end of your message.
For a very short program that hardly does anything interesting that seems to
have a lot of “magic” keywords. But in only a few weeks time you will know what
they all mean, and why they make sense. For now just keep a file that contains the
above basic sample code and copy it every time you want to start a new program
so you do not have to waste time keying in all the junk repeatedly!
3.3.1 Code Layout
Many people have quite strong views about code layout and indentation. That
includes me! The style you will see in these notes almost always places any “}”
vertically below the “{” that it matches. I try to indent anything that is within
such braces by four space positions. Beyond that my guiding principle is to try to
keep my code so that it looks pretty on the screen or page, is efficient in its use of
the page and is as easy to navigate over as I can manage. Saving keyboard effort
is not a high priority, since actually typing in programs is such a very small part
of the total pain that goes into getting a complete and robust working program.
The default emacs idea about indentation and brace layout is differs from mine:
whichever you choose to follow please be consistent and try to make your code
easy for yourself and others to read.
The comment above about efficiency in the use of the page is because when
reading your code it is especially convenient if all the bits you want to see fit
within one screen-full of the editor’s window. Thus I count excessive splitting of
constructs over multiple lines as unhelpful, just as are large swathes of blank lines.
I prefer comments in blocks (which may often make up significant paragraphs)
that describe the code that follows them. And the comments should be readable
English in proper sentences intended to help some poor person faced with revising
or updating the code to correct some imaginary bug or add a new feature.
Java provides some encouragement for special comments that are introduced
with the sequence “/**”8 and going on over possibly many lines until the next
“*/”. These are there to support extra software tools that extract those comments
and format them as separate documentation for the program. In this course I will
illustrate that scheme later on.
Well all the above discussion has just left us with a file Hello.java. Unlike
(typical teaching use of) ML, Java expects programs to be processed by a separate compiler before they are executed. This compiler is a program that checks
the syntax and types of the code you show to it, and translates from the human8 Ordinarily
as well as “//” comments that just run to the end of the line you can write long
comments starting with “/*”.
CHAPTER 3. INTRODUCTION
34
readable9 source file such as Hello.java into a more compact10 digested binary
file (called Hello.class in this case) that can subsequently be executed repeatedly without any need to re-do all the potentially time-consuming checking. To
carry out this conversion you need to say
javac Hello.java
The javac command tends to be jolly taciturn unless it finds something in
your program that offends it. It does not say anything and so after it has run you
may like to use ls to verify that the file Hello.class has been created. Finally
we can run it:
java Hello
Note that when javac was used to compile the program it was essential to
quote the .java extension, while when the program was to be run you must not
use the .class extension that the pre-digested version of the program was given.
This level of apparent inconsistency is not at all restricted to Java, and the exact
rules on matters such as this are liable to differ between different vendor’s sets of
Java tools. What I describe here relates just to Sun’s SDK!
3.3.2 Emacs
The editor emacs is the preferred text editor to use while taking this course. I
think it may be best for most people to start by keeping two windows available
on their screens, one the emacs edit window and the second a command-prompt
from which they can issue the build and run commands directly. When working
with an edit and a command window note that you have to go “File/Save Buffer”11
to get emacs to ensure that the file on disc is brought up to date with respect to the
version you have been editing in its buffer. Provided you do this before issuing the
javac command from your other window it is reasonable and most convenient to
keep emacs loaded throughout your session. It is also possible to compile and run
Java (or other) programs while remaining entirely within emacs, and to get any
reports of syntax errors generated by a compiler to re-position the editor’s caret
close to where the error was detected. But for the rather small programs you will
be working with during this Part IA course all is excessive and using one window
to edit and one to compile as in Figure 3.5 remains simplest.
The next program to be shown is a rather simple extension of the one we have
already discussed, but instead of just printing a fixed message it prints a table of
squares. In a file called Squares.java you should place:
9
Well, at least it is readable if you include enough comments!
for really tiny programs like this one the binary file may be bigger than the source it
relates to, but for and program big enough to be interesting what I say will hold true.
11 Or the equivalent keyboard sequence, Ctrl-x Ctrl-s.
10 Actually
3.3. A COOK-BOOK KICK-START
35
public class Squares
{
public static void main(String[] args)
{
for (int i=0; i<10; i++)
{
System.out.printf("The square of %d is %d%n",
i, i*i);
}
}
}
There are two new things here. The first is the iteration statement
for (int i=0; i<10; i++) { ... }
which arranges to declare a variable called i and set it first to 0, then to 1, then 2
and so on for so long as i<10 remains true. The curious syntax i++ is inherited
from the C programming language and means “increment i”: a less cryptic way of
achieving the same effect would be to write “i=i+1” instead. The single = sign in
Java is an assignment operator and changes the value of the variable named on its
left. The word int is short for “integer” and specifies the type that i should have.
Type Java type int denotes integers which are explicitly limited to a range that
is consistent with representation as 32-bit values. Unlike ML Java expects you to
specify the type of pretty well everything you mention, and when you introduce a
new variable you can change its value later using a = operation without having to
worry about any special extra works like ref.
The second new feature is the string argument to printf where emdedded
percent signs stand for where the numeric values you want displayed need to be
substituted in. The %d indicates that what you want displayed is expected to be an
integer: other letters could be used when you were needing to print other sorts of
item. Once again I need to make a remark this year that is to do with the transition
to Java 1.5: in previous years and in many books you will see this code written as
System.out.println("The square of " + i + " is " + (i*i));
where the plus signs in fact indicate string concatenation and Java is converting
integers to printable form fairly automatically. I prefer the use of printf because
the %d indicates very explicitly that I am about to print an integer (not some other
sort of thing). It can also be extended to give me quite refined control over the
layout of the table I generate.
Note that when I came to want to type in the Squares program to check it I did
not type it in from scratch. Instead a copied the earlier Hello program and adjusted
the few lines in its middle to perform the new operations. Typically it will also
be necessary to change a few comments to make them relate to the new reality,
but creating new code by making incremental extensions to old is a very useful
CHAPTER 3. INTRODUCTION
36
technique and can save a lot of time and effort. It also means that remembering
all those boring bits is at least slightly less necessary.
One further development of the Squares example will illustrate a few more
Java idioms. This code (which I will put in a file Powers.java) computes powers
and does so by a repeated-squaring technique that may be familiar from the MLbased course last term:
public class Powers
{
public static void main(String[] args)
{
// I will use println for simple fixed text
System.out.println("Table of powers");
for (int i=0; i<10; i++)
// .. and printf to incorporate values within a template
{
System.out.printf("%dˆ%d = %d%n", i, i,
power(i, i));
}
}
static int power(int x, int n)
{
if (n == 0) return 1;
int y = power(x, n/2);
if ((n % 2) != 0) return x*y*y;
else return y*y;
}
}
which produces the results
Table
0ˆ0 =
1ˆ1 =
2ˆ2 =
3ˆ3 =
4ˆ4 =
5ˆ5 =
6ˆ6 =
7ˆ7 =
8ˆ8 =
9ˆ9 =
of powers
1
1
4
27
256
3125
46656
823543
16777216
387420489
The new features shown here are the definition of a function and calls to it.
Observe that the types of the arguments for the function and the type of its result
are all explicitly given (as int here). The code does distinctly more arithmetic,
3.3. A COOK-BOOK KICK-START
37
where +, -, * and / stand for addition, subtraction, multiplication and division.
The percent sign % gives a remainder. Numeric comparisons are written with >,
<, >= and <= for the obvious comparisons, and the rather less obvious == for an
equality test and != for inequality.
Conditional statements appear as
if (condition) statement
or
if (condition) statement
else statement
Note that the parentheses around the condition are part of the Java syntax
(inherited from C) and they may not be omitted.
You need to use the word return explicitly to indicate what value your procedure should hand back.
It is a very common beginner’s error to get mixed up about where braces and
semicolons are needed — and mix-ups on this front can cause special trouble
with the else after an if statement. In doubt just remember that you can group
several statements (or indeed just one) together to make a single big statement
just by enclosing them (or it) in braces “{ . . . }”. The braces I have around the
call to printf just after the for were put in not because they are essential (the
call to printf counts as a single statement and could be the thing that the for
loop performed in a repetitive way) but because I think the braces there make it
easier to see just what the range of the for is. Similarly it is often good style to
use braces that are in some sense redundant after the keyword if just to ensure
that the structure of your code is utterly evident to any reader.
The series of small examples above show enough of Java that they can form the
basis for exercises that use integer arithmetic and a few recursive sub-functions.
With luck they contain enough examples of usage that you can now go away and
write all sorts of little programs that perform calculations with at most minor
recourse to the textbook to check exact details.
3.3.3 Drawing to a window: JApplets
I will therefore move onto another cook-book example which shows a different
sort of Java program. The ones seen so far are refereed to as stand-alone applications. The next one will be described as an “applet”. It has an even higher load
of mumbo-jumbo to surround the small bits that are its essential core, but illustrates how you can start to use Java for graphics programming and to interact with
windows, mice and the like. As with my Hello program I will start by quoting
38
CHAPTER 3. INTRODUCTION
Figure 3.7: emacs on Windows, with the “Global Font Lock” option for syntax
colouring.
the important bit of the code that lives in the middle. In this case it will arrange
to keep track of where your mouse last was when you pressed its button, and will
respond to new mouse clicks by drawing a straight line on the screen to join the
old to new position.
Since at this stage I want to make this key part of the code look as short and
easy as possible I have omitted any comments — after all I am about to give an
explanation here in the accompanying text!
3.3. A COOK-BOOK KICK-START
39
In the Hello program we used a function called printf by referencing it relative to some library object System.out. Here we need to suspend disbelief for a
short while and image two things, one called e that allows us to call library functions that reveal the position of the mouse (getX and getY) and another called
g that is analogous to System.out but which supports a function drawLine for
putting a straight line up on the screen. Suppose furthermore that there are integer
variables lastX and lastY that will be used to store the previous position where
the mouse was clicked. It now makes sense to show the kernel of the drawing
program:
int x = e.getX(), y = e.getY();
g.drawLine(lastX, lastY, x, y);
lastX = x;
lastY = y;
Look under the link Java Platform Core API on the web-browsable documentation. Clicking at the top of the screen through Index makes it almost as
quick to look up getX, getY and drawLine as it would be to check for them
in the index of a book. In either case you are liable to find near their documentation the explanation of other related functions, such as drawRect, drawOval,
fillArc, drawString and many many more.
Once one has sorted out how to use one of these in general the rest follow on
naturally, so it can be useful to browse the documentation occasionally to make
yourself familiar with the collection of operations that are supported.
The next natural question is one of where the mysterious e and g came from,
and how it could be arranged that the above code is activated every time the
mouse button is pressed. Well just as a simple stand-alone application has a special function called main, one that deals with the mouse will have one called
mousePressed. This gets the object e passed down to it from the system. all
one needs to know is that the type used to declare this variable is MouseEvent.
Access to the screen is obtained by declaring g to be of type Graphics and initialising it with the value returned by a call to getGraphics12 . These types and
conventions are to some extent part of a large design that underlies the Java libraries, but at this stage the only proper way to cope with them is to copy them
carefully from existing working programs and check details in the documentation.
When you look at the documentation I expect your main initial response to be one
close to “Wow” as you see just how many types and functions Java provides you
with. Overall there is more complexity and power in these libraries than there is
in the language itself. Anyway here is the full version of the mouse click handler
function — not too messy provided one is happy to take the library calls on trust!
12 In
this initial example I use getGraphics but often the object you want will come to you
in other ways.
40
CHAPTER 3. INTRODUCTION
public void mousePressed(MouseEvent e)
{
// I have to obtain access to a drawing context
Graphics g = getGraphics();
// I also need to extract (x,y) co-ordinates from
// the mouse event.
int x = e.getX(), y = e.getY();
g.drawLine(lastX, lastY, x, y);
lastX = x;
lastY = y;
}
I can now give the whole file Draw.java which includes the above important
function definition, but which also has the relevant junk that is needed to connect
it in to the Java run-time environment. You will see that I have this time used
comments from /* to */ for some of the big block comments. The arrangement
with columns of vertical stars is purely a convention that I like and which makes
the range of the comment clearly visible. The lines starting import arrange for
convenient access to several extra Java libraries. You will find import statements
at the top of most of my sample programs from now on and the exact list of things
you need to “import” will seem jolly mysterious. All I can say at this stage is
that you can start by copying the lines I give and that in a week or so you will
understand how to check the Java on-line documentation to sort out exactly what
you need exactly when.
The qualifications (extends and implements) on the declaration of the
Draw class ensure that this program can draw to the screen and respond to the
mouse. When a file contains a class that extends JApplet the rules for it starting up are not like ordinary programs. Instead of defining main it defines the
functions shown here.
/*
A C Norman
* Draw.java
*
Simple applet to draw lines on a screen
*
in response to mouse clicks. See also "Draw.html".
*
/
*
/*
* At the start of almost any Java program it will
* be necessary to incant a few "import" statements to
* provide Java with more convenient access to various
* standard libraries.
*/
3.3. A COOK-BOOK KICK-START
import javax.swing.*;
import java.awt.*;
import java.awt.event.*;
public class Draw extends JApplet
implements MouseListener
{
private int lastY = 0, lastY = 0;
public void init()
{
// I need to activate the mouse event handlers.
this.addMouseListener(this);
}
/*
* Each time the mouse button is pressed I will draw a
* line on the screen from the previous mouse position
* (or (0,0) at the start) to where the mouse now is.
*/
public void mousePressed(MouseEvent e)
{
// I have to obtain access to a drawing context
Graphics g = getGraphics();
// I also need to extract (x,y) co-ordinates
// from the mouse event.
int x = e.getX(), y = e.getY();
g.drawLine(lastX, lastY, x, y);
lastX = x;
lastY = y;
}
/*
* The full mouse event model uses the four extra
* procedures shown below. To keep this code as short
* and simple as I can I will not cause them to do
* anything, but the Java event handler scheme demands
* that they exist. Hence these definitions of functions
* that do nothing at all!
*/
public void mouseReleased(MouseEvent e) {}
public void mouseClicked(MouseEvent e) {}
41
CHAPTER 3. INTRODUCTION
42
public void mouseEntered(MouseEvent e) {}
public void mouseExited(MouseEvent e) {}
}
/* end of Draw.java */
Without a prototype such as the above to start from it could take a huge amount
of reading of the manuals to find all of the Java types and functions to put together.
However once you have the prototype to work from there is at least some chance
that variations on the theme can be constructed by making incremental changes,
and the details of these changes can be sorted out by looking in the manuals close
to the place where the features that are currently used get documented. The version I have given here uses the class JApplet while you may find some books use
Applet (without the initial letter J) that is an older version of something similar.
3.3.4 HTML and appletviewer
The earlier examples were run using commands such as java Hello. This one
is not a stand-alone application but a JApplet, and so has to be run using a thing
called appletviewer. What is more it needs yet another file to be prepared: one
that will let it know how large an area of the screen should be set aside for the
drawing to appear in. This new file must be called Draw.html, and its contents
are as follows:
<HTML>
<BODY>
<APPLET code="Draw.class" width=400 height=400>
Java is not available.
</APPLET>
</BODY>
</HTML>
This file consists of a set of nested sections, where the start of a section is a
word contained in angle brackets and the corresponding end-marker is the same
word but with “/” in front of it. Once again the most interesting part is in the middle where the APPLET tag is used to provide a reference to the compiled version of
our program (ie Draw.class) and to specify the width and height of the window
in which it is to work. The text “Java is not available” should never appear when
you use this file! It is there so that it can be displayed as an error message if this
HTML13 file is inspected using software that does not understand Java. For the
purposes of this course the only use you will make of the file is to say
13 Hypertext
Mark-Up Language.
3.3. A COOK-BOOK KICK-START
43
appletviewer Draw.html
which should cause a 400 by 400 (pixel) window to appear within which you
can click the mouse to good effect. The window that appletviewer should
provide a pull-down menu that contains an entry quit or close than can be
used to terminate the program. Observe (with quiet gloom) that appletviewer
demands that you quote the .html suffix, and that inside the HTML file you have
to specify the full name of your class file (ie including the .class suffix), while
to run simple stand-alone Java applications you just gave the base part of the filename. Ah well!
The Draw program shown here is a useful prototype, but the most glaring
problem it exhibits is that if you rearrange your windows while using it so as to
obscure part of what you have drawn then that bit does not get re-painted when
you reveal the window again.
When you move towards larger programs (ones spread over very many files)
you will probably need to read up about a tool called jar and find out (it is easy!)
how you can package many Java class files into a single archive, and how HTML
files refer to such archives. I will not explain that in this Part IA course.
3.3.5 Exercises
Tickable Exercise 2
Do both parts A and B please.
Part A
The following Java function is a rather crude one for finding a factor of a
number n and returning it, or returning 1 if the number is prime.
static int factorof(int n)
{
int factor = 1;
for (int i=2; i*i<=n; i++)
{
if (n % i == 0) factor = i;
}
return factor;
}
The code works by checking each possible factor from 2 upwards, stopping
when the trial factor exceeds the square root of the number being tested.
This stopping condition is reasonable because if a number n is not prime
√
than it must have at least one factor in the range 2 . . . n.
CHAPTER 3. INTRODUCTION
44
Write a stand-alone Java program that incorporates the above code and that
prints out a list of all the prime numbers from 2 to 100.
Optional: Change factorof to make it more efficient by first letting it
check whether 2, 3 or 5 is a factor and then instead of trying all possible
factor up to the square root of n let it just try those that are 1, 7, 11, 13, 17,
19, 23 or 29 mod 30. Ie do not bother with numbers that are themselves
divisible by 2, 3 or 5. This should lead to making just 8/30 = 26.66% of the
number of test divisions that the original version did. Does the new version
run faster, and if so by about what factor?
Part B: Binomial Coefficients
This exercise is a deliberate incitement to write a very inefficient program.
Later on there will be an example that prompts you to write a much faster
program that can computer the same answers! The binomial coefficients14
may be defined by the rules
n
Cr = n−1Cr +n−1 Cr−1
n
C0 = 1
n
Cn = 1
This definition could naturally turn into an ML function definition
fun binom(n, r) =
if r = 0 orelse r = n then 1
else binom(n-1, r-1) + binom(n-1, r);
Write the corresponding Java code and use it to tabulate 2nCn for n from 0
to 12.
If you are using Java on Unix you should go “time java Binom” to run
the example and when your program has run you will get a report of how
long the computation took. Keep all the output so you can compare both
results and timing data with the method described later on.
(End of tickable exercise)
14 There
is some question as to whether I should
will not confuse you too much!
use the notation nCr
here or
n
r
. I hope this
3.3. A COOK-BOOK KICK-START
45
A better drawing program
The drawing program as presented is very clumsy. There are a number of ways it
could be improved. The suggestions made here are not the correct route towards
a properly finished professional quality drawing program but may still count as
useful practise with Java.
1. In the 400 by 100 window, interpret mouse clicks in the top 50 pixels as
button activity that can select options. The x co-ordinate may be split into
(say) 4 ranges to give four buttons. In mousePressed add:
if (y < 50)
{
if (x < 100) .. action1
else if (x < 200) .. action2
else if (x < 300) .. action3
else .. action4
}
else
{
normal mouse processing
}
2. The crude buttons as above could select whether further regular mouse
clicks drew lines (as before) or used drawOval to draw circles. Another
button might select a drawing colour using
g.setColor(Color.blue); // or red, black etc
3. My code, which was trying to be as short as possible, did not treat the
first mouse click specially, and so all trails started at (0,0). That should
be changed.
4. drawString(string, x, y) places text in a window at the given position. It could be used to label the “buttons”.
I think that the code that you could potentially achieve here would be pretty good
for this stage in the course!
46
CHAPTER 3. INTRODUCTION
Turtle Graphics
The following code shows some
more new features of Java. It defines a paint method (ie function)
in an applet. The appletviewer arranges that this function is invoked
every time the applet’s window is
uncovered or otherwise needs redrawing, and so it leads to pictures
that are a lot more robust than the
mouse-driven Draw program shown
earlier. The code also uses a new
type, double which is for floating
point numbers, and some calls to the
Maths library to compute sines and cosines. The odd notation (int)x indicates
that the code wants to convert the floating point value x into an integer (int) so it
is of the correct type to be passed on to drawLine.
Put the code in a file Turtle.java and prepare a suitable associated file
Turtle.html. Experiment with the code and see how the image changes depending on the values of the three variables marked. For most values of inc I
seem to find that a closed figure is drawn provided N is large enough, but I have
some trouble producing an explanation of why or a characterisation of exactly
what values of inc will lead to this behaviour. I also find the degree of symmetry hard to explain. Generally this is an illustration of the fact that quite short
programs can have behaviour that is complicated to explain!
/*
A C Norman
* Turtle.java
illustration
of
Turtle
Graphics
and
the
"paint"
method.
*
/
*
import javax.swing.*;
import java.awt.*;
import static java.lang.Math.*;
public class Turtle extends JApplet
{
public void paint(Graphics g)
{
// Try changing the following 3 numbers...
double size = 5.0, inc = 11.0;
int N = 5000;
double x = 200.0, y=200.0,
3.3. A COOK-BOOK KICK-START
47
th1 = 0.0, th2 = 0.0, th3 = 0.0;
for (int i=0; i<N; i++)
{
th3 = th3 + inc;
th2 = th2 + th3;
th1 = th1 + th2;
double x1 = x+size*cos(PI*th1/180.0);
double y1 = y+size*sin(PI*th1/180.0);
g.drawLine((int)x, (int)y, (int)x1, (int)y1);
x = x1;
y = y1;
}
}
}
/* end of Turtle.java */
The code is really using angles in degrees (not in radians), and the variables
th1, th2 and th3 hold values that are angles. As coded above some of these
angles can grow to ridiculously large values, it might make sense to insert lines
based on the prototype
if (th2 >= 180.0) th2 = th2 - 360.0;
in suitable places with a view to keeping all the angles that are used in the range
−180.0 to +180.0.
The import static java.lang.Math.* line makes it possible to use
sin, cos and PI in the simple way shown.
Note that in Java (and indeed with many window systems) the y co-ordinate
starts at 0 at the top of the screen and increases as you go down. This makes sense
(sort of) when the screen is containing text, in that counting lines you normally
start at the top. For pictures it can be a little muddling until you are used to it, and
can mean that things sometimes come out upside down the first time you try them.
48
CHAPTER 3. INTRODUCTION
Chapter 4
Basic use of Java
4.1 Data types, constants and operations
The first section of these notes introduced a few small but complete Java programs,
but when you type them into the computer you still have to take a great deal on
trust. But with those examples to use as a framework I can now start a more
systematic introduction of the Java language and its associated libraries. Actually
in the lectures I expect to skim over this material very rapidly: you will in reality
learn about the Java data types and syntax as you write programs. However I
view it as important that you have reference material in your handout that shows
what everything is so that if you have trouble in your coding you have somewhere
reasonably close at hand where you can check some details. However if you need
a gentle path to settle into Java programming I do suggest that you try various of
the example programs and exercises here so that you get used to and comfortable
with a good range of Java syntax.
4.1.1 Reserved Words
The first thing to do is to catalogue the words that are reserved: if you accidentally
try to use one of these names as the name of one of your variables or functions you
can expect most curious error messages from Java! So even though I do not want
to explain what all of these mean yet it may help you if I provide a list of words
to be avoided. In some cases of course the presence of a word here will alert you
to the availability of some operation, and you can then look up the details in the
manual. A clever editor might display words from this list in some alternative
colour to help you notice any unintentional uses. An even more clever one might
use different colours for the one (such as int) that name basic types, the ones
such as for that introduce syntax and ones like true that are just the names of
49
50
CHAPTER 4. BASIC USE OF JAVA
Figure 4.1: Start with some small examples. . .
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
51
important built-in constants.
abstract
case
continue
enum
float
import
native
protected
strictfp
throw
void
assert
catch
default
extends
for
instanceof
new
public
super
throws
volatile
boolean
char
do
false
goto
int
null
return
switch
transient
while
break
byte
class
const
double
else
final
finally
if
implements
interface
long
package
private
short
static
synchronizedthis
true
try
A joke about the above table of reserved words is that at present Java does not
actually use them all — specifically const and goto do not represent any operation supported by Java. By prohibiting people from using these words as names
for variables the Java designers have left themselves a little flexibility to extend
the language in the future if they want to or are forced to, and they can perhaps
give better error messages when people who are used to some other language that
does include these keywords first tries out Java.
There are some other words which are used for names that perform important
system functions. If you unintentionally define a function with one of these names
you might find that you have introduced side-effects of amazing magnitude when
the system calls your function instead of the one it was expecting! Beware here,
because although incorrect use of a genuine reserved word will result in a syntax
error it could be that defining a function with one of the following names would
have subtle or delayed bad consequences rather than a nice clean instant crash.
clone
notify
equals
notifyAll
finalize
toString
getClass
wait
hashCode
OK so the above information is more negative then positive, but I hope it will
rescue a few of you from otherwise most mysterious behaviour when you might
otherwise have tried to use one of the reserved words for your own purposes.
4.1.2 Basic Types
Earlier examples used the word int to declare integer variables, and the range of
values that can be represented goes from around −2 billion to around +2 billion.
To be more precise the smallest valid int is −231 = −2147483648 and the largest
one is 231 − 1 = 2147483647. You are not expected to remember the decimal form
of the numbers, but you should be aware of roughly how big they are. Integer
CHAPTER 4. BASIC USE OF JAVA
52
overflow is ignored: the result of an addition, subtraction or multiplication will
always be just what you would get by representing the true result as a binary
number and just keeping the lowest 32 bits. A way to see the consequences of this
is to change the Powers program so it goes up to higher powers, say 20. The final
section of the output I get is
...
8ˆ8 =
9ˆ9 =
10ˆ10
11ˆ11
12ˆ12
13ˆ13
14ˆ14
15ˆ15
16ˆ16
17ˆ17
18ˆ18
19ˆ19
16777216
387420489
= 1410065408
= 1843829075
= -251658240
= -1692154371
= -1282129920
= 1500973039
= 0
= 1681328401
= 457441280
= -306639989
where the value shown for 1010 is clearly wrong and where we subsequently get
values that probably count as rubbish. Note both the fact that overflow can turn
positive values into negative ones (and vice versa) and the special case (obvious in
retrospect) where 1616 shows up as zero. Since 16 is 24 the binary representation
of 1616 is clearly a 1 followed by a string of 64 zeros, and in particular the least
significant 32 bits are all zero. This lack of detection of integer overflow is sometimes convenient but it is also a potential major source for getting wrong answers
without even knowing it.
Java provides several alternative integer-style primitive data-types which represent different trade-offs between expected speed, space and accuracy. They are:
byte: 8-bit integers in the range −128 to +127;
short: 16-bit integer, range −215 = −32768 to 215 − 1 = 32767;
int: 32-bit integers as discussed already;
long: 64-bit integers, is range is from −263 to 263 − 1 which means that almost
all numbers with up to 19 decimal digits can be represented.
It may be helpful to those who are not already used to the binary representation
of signed values if I tabulate the representation used for the byte datatype. The
wider integral types use just the natural generalisation:
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
Number Representation in binary
−27 26 25 24 23 22 21
127
0 1 1 1 1 1 1
126
0 1 1 1 1 1 1
...
3
0 0 0 0 0 0 1
2
0 0 0 0 0 0 1
1
0 0 0 0 0 0 0
0
0 0 0 0 0 0 0
−1
1 1 1 1 1 1 1
−2
1 1 1 1 1 1 1
−3
1 1 1 1 1 1 0
−4
1 1 1 1 1 1 0
...
−126
1 0 0 0 0 0 1
−127
1 0 0 0 0 0 0
−128
1 0 0 0 0 0 0
53
20
1
0
1
0
1
0
1
0
1
0
0
1
0
One way to understand how the negative numbers arose is to see that −1 is the
bit-pattern that has the property that if you add 1 to it and ignore the final carry
you get the representation that means 0. It can also help to suppose that negative
number “really’ have an infinite string of 1 bits glued onto their left hand end. The
representation used is known as “two’s complement”.
When you write an integer literal in your Java code you can write it in decimal,
octal or hexadecimal (base-16). You can also make your written integer either of
type int or type long; there is no direct way to write either a byte or short.
Decimal numbers are written in the utterly ordinary way you would expect. You
add a suffix “L” if you want to make the value a long and you should always do
this if the value is outside the range of ordinary available to an int, but you might
sometimes like to do it for even small values when using them in a context where
arithmetic on them should be done in long precision. Examples are:
12345
1234567890123L
10L
1000000L*1000000L
1000000*1000000
an ordinary int
a long value
a long but with a smallish value
an expression where the L suffix
matters
without the L this would overflow.
My belief is that hardly anybody ever wants to write a number in octal these
days1 , but Java allows it, taking any number that starts with 0 as being in octal.
1 But
this may be just a matter of fashion, and perhaps elsewhere in the world octal is still
appreciated.
CHAPTER 4. BASIC USE OF JAVA
54
Thus 037 is the octal way of writing the number 31. The L suffix can be used to
specify long octal values. Observe a slight jollity. If you write the number 0 it
is interpreted as being in octal. Fortunately zero is zero whatever radix is used to
write it!
Hexadecimal is much more useful. Each hexadecimal digit stands for four
bits in the number. Letters from A to F are used to stand for the digits with weight
10 . . .15. Hexadecimal numbers are written with the prefix 0X. Note that the suffix
L for long, the X in hexadecimal numbers and the extended digits from A to F can
all be written in either upper or lower case. I strongly recommend use of upper
case for L since otherwise it is painfully easy for a casual reader to muddle 10l
(long ten) and 101 (one hundred and one).
Here are some numbers written in hexadecimal
0X0
0xe
0xffffffff
0xBadFace
0x7fffffff
0x80000000
0x00010000
0x7fffffffffffffffL
this is zero, not gravy powder
otherwise 14
-1 as an int
what other words can you spell?
largest int
most negative int
2 to the power 16
largest long
I rather suspect that the main importance of byte and short is for when you
have huge blocks of them2 where the fact that they take up less space can be of
practical value.
Floating point values also come in two flavours, one with a larger range and
precision than the other. The more restricted one is called float. A float uses
32-bits of memory and can represent values up to about 3.4e383 with a precision
of six to seven significant figures. Until you have sat through the course on numerical analysis please avoid use of it4 . The more sensible floating point type is called
double and uses 64 bits to store numbers with magnitude up to about 1.7e308,
with an accuracy of sixteen or seventeen significant figures. The internal representation of floating point values and the exact behaviour in all circumstances was
originally taken from an International Standard referred to as IEEE 754. Some
bit-patterns are reserved to represent “+∞” and “−∞” while others are values that
are explicitly not representations of valid floating point values — these are known
as NaNs (Not A Number). A few possibly unexpected effects arise from this. For
2 See
the description later on of arrays.
3.4 × 1038.
4 In fact a number of the Java library functions require arguments of type float, so it is not
possible to avoid this type. Its use is satisfactory in circumstances where the precision it supports
is all that is justifiable, for instance when specifying the brightness of a colour.
3 ie
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
55
instance floating point division never fails: 0.0/0.0 yields a NaN, while any other
value divided by 0.0 results in an infinity. Also if u is a floating point value then
it is possible for the expression u == u to evaluate to false (!) because the
rules for all numeric comparison operators is that they return false when at least
one of their arguments is a NaN. Another oddity is that one can have two floating
point values u and v such that both u == v and (1.0/u) != (1.0/v)! This
oddity is achieved by having u = -0.0 and v = +0.0. These delicacies are almost certainly exhibited by most other languages you will come across, but Java
documents them carefully since it is very keen indeed to make sure that Java program will give exactly the same results whatever computer it is run on. Even if
it does very delicate things with the marginal and curious cases of floating point
arithmetic. Recent versions of the Java language use a keyword strictfp to
indicate places where all the consequences of IEEE floating point must be honoured: specifically its use means that the results computed should be identical
whatever machine run on, and will have rounding errors exactly as expected.
Without strictfp and on some computers Java will deliver results that are both
more accurate and are computed faster!
Here are some floating point constants:
0.0
0.0F
1.3e-11
22.9e11F
22.9e11D
1e1
2.
.2
2D
this is double (by default)
if you REALLY want a float
specify an exponent (double)
float not double
be explicit that a double is required
no "." needed if there is an "e"
"." can come at end
"." can come at start
must be a double because of the D
I would suggest that you always make your floating point constants start with
a digit and contain a decimal point with at least one digit after it since I think that
makes things more readable.
In Java the result of a comparison is of type boolean, and the boolean constants are true and false. As in ML (and unlike the situation in C, in case you
know about that), boolean is a quite separate type from int.
Despite the fact that this section is about the Java primitive types and not about
the operations that can be performed on data, it will make some of my examples
easier if I next mention the ways in which Java can convert from one type to
another. In some cases where no information will be lost (eg converting from a
byte or short to an int) the conversion will often happen without you having to
worry much about it and without anything special having to be written. However
the general construction for making a type conversion is called a cast, and it is
written by using a type name in parentheses as an operator. We have already
CHAPTER 4. BASIC USE OF JAVA
56
already seen a couple of examples in the Draw program where (int)x was used
to convert the floating point value x into an integer. The opposite conversion can
then of course be written as in
for (int i=1; i<10; i++)
System.out.printf("%22.8g%n", 1.0/(double)i);
where the (double) is a cast to convert i to floating point5. The format specifier6 %22.8g is for printing a floating point value in a General format using a
precision of 8 significant figures and padding with blanks to make 22 characters
printed in all. Until you understand exactly when automatic conversions apply it
may be safest to be explicit. Java allows you to write casts for those conversions
that it thinks are sufficiently reasonable. You can cast between any of the flavours
of integer. When you promote from a narrower integer to a wider one the value
is always preserved. When you cast from a wider integer to a narrower one the
result is what you get from considering the binary representation of the values
concerned, and the cast just throws away unwanted high-order bits. Casts from
integers to float and double preserve value as best they can7. Casts from floating point values to integers turn NaNs into 0, and infinities into either the most
positive or most negative integer. Floating point values that are too large to be
an int or long also turn into the largest available integer. The exact rules for
casts from floating point values to byte and short are something to look up in
the reference manual in the improbable case it matters to you. There are no casts
between boolean and other types. You need to use explicit expressions such as
(i != 0) to map from an integer to a truth-value8.
The type char can be used to declare a variable that holds a single character.
To write a constant suitable for putting into such a variable you just write the
relevant character within single quote marks, as in
char mychar = ’A’;
if (myChar == ’q’) ...
It is frequently necessary to use characters that do not fit so neatly or clearly
between quotes. For instance the single quote character itself, or a “character” to
represent the end of a line. A set of escape codes are used for these, where instead
of a single character one writes a short sequence starting with the escape character
“\”. The supported escape sequences are:
5
In this case the cast is not needed: Java will do the required conversion so that it can perform
the division.
6 You will see a bunch of common format specifiers just in examples here. You can look up full
details in the on-line documentation, or find a medium-sized synopsis later in these notes.
7 Casts from int or long to float or from long to double can not always preserve
an exact result because the floating point format may not have enough precision available. The
closest floating point value to the true result will be delivered.
8 Unlike the position in C where there is not much distinction between integers and booleans.
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
\n
\"
\’
\\
\b
\t
\f
\r
57
newline, linefeed (very commonly used)
double quote mark
single quote mark
use when a single \ is wanted
backspace
tab
newpage, formfeed (unusual)
carriage-return
Carriage returns are used in Windows text files and as line separation in some
Internet protocols, but when creating simple text files you do not generally see
or need to mention it: Java does any necessary conversions to you so that on
Windows, Macintosh and Unix an end of line in a file is talked about in your
programs as just ‘\n’.
In addition it is possible to write \nnn where nnn stands for 1, 2 or 3 octal
digits: this indicates the character with the specified character-code in a standard
encoding. Use of octal escapes is not at all common. Furthermore Java allows
inclusion of characters from an astonishingly large character set by use of a notation \u followed by four hexadecimal digits. The 16-bit number represented by
the hexadecimal digits is taken as being in a set of character encodings known
as Unicode. Casts between int and char give direct access to this encoding.
For example \u2297 and (char)0x2297 both give the character “⊗”. In fact
the Unicode escapes do not just apply within Java character literals but can be
used anywhere in a Java program where you want an unusual symbol — and
this means that in some sense you can have variables names with Greek, Russian and Eastern glyphs in them. Unicode gradually becoming more widely used,
but most computers still do not have full Unicode fonts installed, and so exotic
characters will not always be displayed properly even though within Java they are
handled carefully. The following applet displays the characters that are available
using the viewer it is run under. It uses a cast (char) to convert an integer to
a character and some fresh library calls (eg setFont(new Font(...)) and
drawString). It also illustrate something that you will probably want to retrofit
to most of the little examples in these notes. It allocates a BufferedImage that
it draws into, and then the paint method just displays whatever is in the bitmap.
This does wonders for arranging that when you obscure bits of your window the
content gets re-painted nicely!
It also makes a crude modification of the earlier Draw program so that mouse
clicks at various places in the window adjust the range of characters displayed.
/*
*
*
Unicode.java
A C Norman
CHAPTER 4. BASIC USE OF JAVA
58
* Display the Unicode characters as supported
* by the current browser.
*/
import
import
import
import
java.awt.*;
java.awt.event.*;
javax.swing.*;
java.awt.image.*;
public class Unicode extends JApplet
implements MouseListener
{
private boolean isFilled = false;
private int fontSize = 20; // or whatever!
private int page = 0;
private BufferedImage p =
new BufferedImage(
32*fontSize,
35*fontSize,
BufferedImage.TYPE_BYTE_BINARY);
public void init()
{
addMouseListener(this);
}
public void mousePressed(MouseEvent e)
{
if (e.getX() < 200) page++;
else page--;
if (page > 63) page = 0;
if (page < 0) page = 63;
isFilled = false;
repaint(); // force screen to re-draw
}
public void paint(Graphics g)
{
if (!isFilled) fillImage();
// Note drawImage may need repeating!!!
while (!g.drawImage(p, 0, 0, this));
}
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
59
void fillImage()
{
Graphics g = p.getGraphics();
g.setColor(Color.WHITE); // background
g.fillRect(0, 0, 32*fontSize, 35*fontSize);
g.setFont(new Font("Serif",
Font.PLAIN, fontSize));
g.setColor(Color.BLACK); // text
g.drawString("page = " +
Integer.toHexString(32*32*page),
0, fontSize);
for (int y=0; y<32; y++)
{
for (int x=0; x<32; x++)
{
char c = (char)((32*page+y)*32+x);
g.drawString(String.valueOf(c),
fontSize*x,
fontSize*(y+2));
}
}
isFilled = true;
}
public
public
public
public
void
void
void
void
mouseReleased(MouseEvent e) {}
mouseClicked(MouseEvent e) {}
mouseEntered(MouseEvent e) {}
mouseExited(MouseEvent e) {}
}
/* end of Unicode.java */
The output from this program will depend on the range of fonts installed on
the computer you run it on. PWF Linux has a range of European characters, mathematical symbols and oddments available. While preparing these notes I ran the
code on my home Windows XP system where all sorts of fonts have accumulated
over the years, and the image included here (figure 4.2.) is from there. I also
use a program called Vmware which lets me install many “virtual” computers on
my single home one: using that I installed essentially the version of Linux used
on the PWF but told the Linux installer to include support for all available languages: by moving some files into a directory “jre/lib/fonts/fallback” I
could get results very similar to those that I get from Windows. A message I hope
you will absorb here is that Java itself provides portable support for international
and special-purpose character sets you may need to configure its runtime before
you can take full advantage of it. Also before you distribute applications relying
60
CHAPTER 4. BASIC USE OF JAVA
Figure 4.2: Unicode characters in the range 0x3000 to 0x33ff
on that you have to concern yourself with how well your customers’ operating
systems will deal with the fonts!
We have already seen string literals in our code, just written within double
quote-marks. The associated type is String. Although the use of capitals and
lower case is just a convention in Java the fact that the type is String rather
than string is a hint that this does not have exactly the same status as the types
int, char and so on. In fact String is the name of a quite complicated datatype (in Java we will find that this is known as a class) and this class pro-
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
61
vides access to a number of string conversion and manipulation functions. We
have already seen “+” can be used to give string concatenation. Java also arranges that if one argument for + is a String it will take steps to convert the
other to String format so that this concatenation can take place. You can look
up “Class java.lang.String” in the on-line documentation or a reference
manual to see that there are standard library functions for case-conversion and all
sorts of other string operations.
For now the points to observe are that
1. Strings are represented by a data-type that exports functions to find the
length of a string, concatenate strings and perform various conversions;
2. Strings are read-only, so if you want to change one you in fact make a new
string containing the adjusted text;
3. Strings are not the same as arrays9 of characters;
4. It is not necessary to memorise every single string operation that Java provides.
Java supports arrays. An array is just a block of values where one has the
ability to use an integer index to select which one is to be referenced. The types for
arrays are written with the array size in square brackets. An empty pair of square
brackets means that the size is not being specified at that point, but it will be made
definite somewhere else in the program. We saw an array declaration as early as
the Hello program where the function main was passed an array of Strings.
In this case the array will be used to pass down to the Java application any words
given on the command line after the parts that actually launch the application:
// File "Args.java"
//
Display arguments from command line
public class Args
{
public static void main(String[] args)
{
for (String s : args)
System.out.println(s)
}
}
This introduces a new version of the for statement. It can be compiled and then
run by saying
9 Which
will be covered in the next section of these notes!
CHAPTER 4. BASIC USE OF JAVA
62
javac Args.java
java Args one two three
it prints out
one
two
three
The points to notice here are that the type of argument that main was an array
of strings, and the for loop will obey its body once for each string in that array.
An alternative and older-fashioned way of achieving the same effect would be to
find the length of the array and count, indexing into the array to extract values
explicitly:
for (int i=0; i<args.length; i++)
System.out.println(args[i])
In this case the array held Strings, but Java arrays can be declared in forms to
hold any sort of Java data. This includes having arrays of arrays, which is the Java
way of modelling multi-dimensional structures.
In Java a distinction is made between declaring a variable that can hold an
array and actually creating the array that will live there. Declaring the variable
happens just as for declaring integer or floating point variables, and you do not at
that stage specify how big the array will be:
{
int[] a;
...
has declared10 a to be ready to store an integer array (of unspecified size and
currently unknown contents), much as
{
double d;
...
says that d will subsequently be able to store double-precision floating point values. There are two ways that the actual concrete array can come into existence.
The first is to combine the declaration with an initializer that makes the array and
fills in its elements:
10 The
syntax that I will try to use throughout these notes has all declarations written as a
type followed by the name of the variable that is to be declared. Thus int[] is the type of
an array able to hold integers. When you declare a variable of an array type Java allows you
to put the brackets either next to the base type (as I will generally do) or after the name of the
variable that is being declared, as in int myArray[];. This latter case is perhaps useful
when you want to declare a bunch of variables at once, some scalars and some arrays, as in
int simple,row[],grid[][];.
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
{
63
int[] p = {2,3,5,7,11,13,17};
...
where the values within the braces can in fact be arbitrary (integer-valued)
expressions. The second way of creating an array uses a keyword new which
is Java’s general mechanism for allocating space for things. The word new is
followed by a type that describes the structure of the item to be allocated.
{
int[] fairly_big = new int[1000];
...
In this case the array contents will be left in a default state. In fact Java is very
keen to avoid leaving anything undefined or uncertain, since it wants all programs
to execute in exactly the same way every time and on every machine, so it demands
that a fresh integer array starts off holding 0 in every element. It has analogous
rules to specify the initial value of any other field or array element that has not
had a value given more explicitly.
Note that all Java arrays use subscripts that start at 0, so an array of length
1000 will accept subscripts 0, . . . , 999. Attempts to go outside the bounds will be
trapped by the Java run-time system. Subscripts must be of type int. You may
not use long values as subscripts. If you write a char, byte or short expression
as a subscript Java will convert it to an int for you as if there had been a suitable
cast visible.
When an array is passed as an argument to a function the called function can
update the members of the array, but if it creates a whole new array by something
such as
args = new String [5];
this will replace the array wholesale within the current function but have no effect
on the calling routine. The terminology that Java uses for all this is that it will
pass a “reference” to the array as the actual argument.
The following example shows the creation of an array, code that fills in entries
in it, a slightly dodgy illustration of the fact that two-dimensional arrays can be
viewed as arrays of one-dimensional arrays and a crude demonstration of how you
might print multiple values of a single line by building up a string that maps the
entire contents of the line.
CHAPTER 4. BASIC USE OF JAVA
64
// Array1.java
// Create a 3 by 3 array, swap rows in it (!)
// and print tolerably neatly.
public class Array1
{
public static void main(String[] args)
{
int [][] a = new int[3][3]; // 3 by 3 array
int [] b;
// array of length 3
for (int i=0; i<3; i++)
// fill in all of a
for (int j=0; j<3; j++) a[i][j] = i+10*j;
// The next line recognises that a[i] are 1-dimensional
// arrays of length 3. It swaps two of them around!
b = a[0]; a[0] = a[2]; a[2] = b;
for (int i=0; i<3; i++)
// Print each row
{
String s = "";
// Build row up here
for (int j=0; j<3; j++)
s = s + " " + a[i][j];
System.out.println(s); // Then print it
}
}
}
which prints
2
1
0
12
11
10
22
21
20
The things to notice in the above example are firstly that variables a and b are
declared with array types, but these types neither specify sizes nor imply that a
genuine array actually exists, and secondly the way in which the two-dimensional
array is dismembered. Observe also the syntax associated with new for allocating
space for the array, and the fact that nothing special had to be done at the end
to discard the space so allocated. Java will recycle memory previously used by
arrays (and indeed any other structures) once it knows that they are no longer
needed. This is of course just like the situation in ML.
We have seen a number of other types in the sample programs. As well as
String there was Graphics, Font and MouseEvent. Java 1.2 defines over 500
such non-simple types! Thus one thing you can be certain of is that I will not
discuss all of them, and neither will the follow-on Java course next year. Each of
these has (in some sense) the same status and possibilities as the programs you
have written where you start off by declaring a new class. Each of String,
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
65
Graphics and so on represents a class and its implementation might well be in
Java stored in a file that starts off
public class Whatever ...
You have seen with the classes that you define for yourself that a class is a
context within which you can define a collection of functions, and so it should be
no surprise that each of the 500+ Java library classes provides a whole bunch of
associated functions (eg for String we have mentioned the valueOf operation,
and for Graphics we have used drawLine and drawString). There are thus
literally thousands of library functions available. Their organisation into classes
provides some structure to the collection, but in the end you probably have to find
out about the ones you need to use by searching through the documentation. These
notes will introduce a sampling of library classes and the functions they support
with a view to directing you towards interesting areas of Java functionality. Very
often I find that the best way to start to understand the use of a new part of the
class library is to study and then try to modify some existing code that uses it.
The Java designers suggest use of a convention where the names of ordinary
variables and functions start with a lower case letter while class-names start with
a capital. They further recommend use of words spelt in all capitals for things that
should be thought of as constants (such as PI that we used earlier). The syntax
associated with declaring something immutable will be covered later on once we
have got through the use of other important words such as public and static
which are of course still unexplained.
4.1.3 Exercises
Tickable Exercise 3
The function System.currentTimeMillis() returns a long value that is the
count of the number of milliseconds from midnight on 1st January 1970 to the
moment at which it is executed. Thus something like
long start = System.currentTimeMillis();
for (int i=0; i<13; i++)
{
System.out.println(binom(2*i, i));
}
long timeSpent = System.currentTimeMillis()-start;
System.out.println("Done in " + timeSpent +
" milliseconds");
in the middle of a program can be used to record how long it takes to run. Note
that this is the time as measured by a stop-watch (or hour glass), and will depend
66
CHAPTER 4. BASIC USE OF JAVA
quite strongly on how many other people are using the computer at the same time.
On a single-user computer it can give a tolerably reliable indication of the cost of
a computation and even on a shared machine it is better than no information at all.
1. Adapt the Binomial Coefficients program suggested in the previous set of
examples so that it reports the time it takes to get as far as displaying 24C12,
which (I think) has the value 2704156. Your submission to the assessors
should include a table of the values of 2nCn for n from 1 to 12, and the
number of milliseconds that your program took to run.
2. Remove the definition of the sub-function that you used to compute the binomial coefficients, and add to your program a line that declares and create
an array called c of size 25 by 25. Set the c[0][0] to 1. Now the first row
of the matrix holds values of 0Cr .
Now fill in subsequent rows one at a time using the rules
c[n][0] = c[n][n] = 1
c[n][r] = c[n − 1][r − 1] + c[n − 1][r]
so that the matrix gradually gets filled up with binomial coefficients. Keep
going until the 24th row has been filled in. Then print out the values of
c[2*i][i] for i from 0 to 12, and again measure how long this takes.
3. [From here on is optional] The above calculation can be done using a onedimensional array, so that at each stage in the calculation it holds just one
row of binomial coefficients, ie values of nCr for a single value of n. At each
stage by filling its values in in reverse order something like
c[i] = 1;
for (int j=i-1; j>0; j=j-1) c[j] = ...
the new values can replace the old ones in such a way that nothing is overwritten too early. The for loop shown here sets j first to the value i-1,
then to i-2, and so on all the way down to 3, 2 and finally 1. I could of
course have written j-- or --j where here I put j=j-1!
Write this version of the program using an array of length 80, and make the
array contain long values rather than just int. First arrange that on every
even row it prints the middle element from the part of the array that is in use,
so it duplicates the output printed by the previous two examples. Then make
the loop continue further and thus find (by inspection) the largest value of i
such that 2iCi can be represented exactly as a Java long. The value is less
than 40.
(End of tickable exercise)
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
67
A Numerical Monster
A very fine paper called “Numerical Monsters” by Essex et al[12] explains how many calculations that
you might think were straightforward have extra depth when done
using finite precision computer arithmetic. One example is the function
y = (x2 − 2x + 1) − (x − 1)2
In ideal arithmetic y would always be
zero. If however the function is computed using floating point arithmetic
as shown (and provided an overclever compiler does not do algebraic re-arrangement of the formula:
the Java compiler is well-behaved in
this respect) an interesting graph can emerge. For instance the graph shown here
was produced using a very simple Java program and plotting for x in the range 13.0e-8<x<1.0+3.0e-8 and y from -1.5e-16 to +1.5e-16. Write your own version
of a program to re-create this graph, and investigate the what it shows near other
values of x. Two cases I will suggest investigating are 1.0e8<x<2.0e8, |y|<10
and 14999.99999999253<X<14999.99999999257, |y|<3.2e-8. Your challenge
is to understand exactly how the finite nature of computer arithmetic leads to the
precise patterns generated, and how these patterns would vary if details of the
arithmetic model were altered.
The Dutch National Flag
Provided that at the start of your program you have written
import java.util.*;
The code
int [] a = new int [1000];
Random r = new Random();
for (int i=0; i<1000; i++) a[i] = (byte) r.nextInt();
first makes an array of length 1000. It then creates a new random-number generator (called r), and finally calls the random number generator 1000 times to fill in
entries in the array. The cast to type byte ensures that each entry in the array will
CHAPTER 4. BASIC USE OF JAVA
68
end up in the range from −128 to +127. There will of course be duplicate values
in the array.
The task you have to achieve is to rearrange the numbers in the array so that
they fall into three bands. The first band, say all the elements from 0 to m, should
contain all the numbers x with x < −40. The second band (m + 1 to n) will be for
−40 ≤ x < 40, while the final band (n + 1 to 999) is for x ≥ 40. This is known as
the Dutch National Flag problem because its originator (E J Dijkstra) presented it
in terms of values that had one of the three colours that his country’s flag11 used,
rather than the numeric ranges I have suggested here.
The problem would probably be easy if you could allocate three fresh arrays
and copy each item from the original into one or the other of these, based on its
“colour”. At the end you could then copy the three chunks back into the original
array at the positions they needed to go. But this challenge involves the idea of
efficiency too, and your final solution must not use any extra arrays, and it should
ideally inspect or move each value as few times as it can. Note that just sorting the
values in the array into ascending order would satisfy the objectives that concern
where values must end up, but since the problem does not state anything at all
about any desired order of the items that fall within any of the three bands a
solution based on sorting is over-elaborate and too expensive.
It may well be useful to try your hand at the Polish Flag problem — my encyclopaedia shows the Polish flag as having just two stripes12. Thus the Polish Flag
problem is to rearrange the values in the original chaotic array so that all negative
ones (say) come before all positive ones, but without any further constraint on the
re-ordering apart that it should be achieved reasonably efficiently.
The Mauritian Flag seems to go Red, Blue, Yellow and then Green. . .
Matrix Operations
Set up two 5 by 5 arrays of type double[][]. Fill in the first so that13 the element
at position (i, j) has value 1/(i + j + 1.0). Fill in the other so that the the elements
on the diagonal have the value 1.0 while all other elements hold 0.0.
The program wanted now will be one that gradually turns the first matrix into
one that has just 1.0 elements down its diagonal and zeros elsewhere. The permitted operations are
1. Multiply all the elements in a row by the same non-zero value;
2. Subtract a multiple of one row from another.
11 Red,
White and Blue in that order
and White
13 This form of matrix is known as a Hilbert Matrix.
12 Red
4.1. DATA TYPES, CONSTANTS AND OPERATIONS
69
and whenever one of these operations is performed on the first matrix it must also
be performed on the second.
The first matrix can be made diagonal by tidying up first the first column, then
the second, then the third and so on. To tidy up column i you first multiply row i
by 1/ai,i , since this will leave14 the element on the diagonal as 1.0. Then for every
row apart from row i you subtract a suitable multiple of row i so as to make the
element in column i vanish.
Do this and display the elements of the second matrix, which should in fact
have ended up as the inverse of the original Hilbert matrix!
Encryption
The following code fragment starts with a string called key and fills out an array
k with repeated copies of the character codes from the key until k has 256 entries
in it:
String key = "My Secret Key";
int keyLength = key.length();
int [] k = new int [256];
for (int i=0; i<256; i++)
k[i] = (int)key.charAt(i % keyLength) % 256;
All the values stored in k have been reduced so as to be in the range 0 to 255.
Observe the use of functions length() and charAt() from the String class. I
have used a fixed string as the keyword here.
The repeated use of the fixed numeric constant “256” in this code is a stylistic
oddity. In some ways once the array k has been declared it might be nicer to use
k.length to talk about the number of elements it has. I took the view when
writing this that the exact size of the array is part of the core specification of the
algorithm I am implementing. . . When you write this program whatever else you
do please do not use your password as text in the program you write!
The program you have to write here may be related to an encryption method
known as RC4 that was once a trade secret of RSA15 but which was published
anonymously and presumably improperly a couple of years ago. RC4 is used
as the encryption technology in a large number of generally used packages and
although its security may not have been proved it is widely believed to be respectable. It is also fast.
14 For
now assume please that the diagonal element was non-zero so that the division behaved
properly and did not end up yielding and IEEE infinity.
15 The major American encryption and security company. You may like the view consideration
of the proper uses of such code as an exercise relating to the Professional Practise and Ethics
course.
CHAPTER 4. BASIC USE OF JAVA
70
The first part of the procedure is to create an array s of 256 integers, and initialise it. It is first set so that the value at position i is just i (i runs from 0 to 255).
Now your code should scramble this up using they key k using the following process:
For i from 0 to 255 do
Let j be (s[i] + k[i]) reduced to 8 bits
Swap the items at positions i and j in the array s
The term “reduced to 8 bits” can be implemented by just taking the remainder16 when the value concerned is divided by 256.
At the end of this the array s holds a working collection of scrambled data.
This is used to generate a sequence of 8-bit numbers which can be combined
with a message to encrypt it. Starting with variables i and j both zero the next
encryption-number is obtained as follows:
Increment i modulo17 256
Set j to j + s[i], again modulo 256
Swap s[i] with s[j]
Let t be s[i] + s[j] modulo 256
The result is s[t]
The algorithm is clearly short enough to be utterly memorable! The sequence
of numbers it generates can be added to the (8-bit) character codes of a message
to give the encoded version, and if the recipient knows the key that was used then
decryption is just generating and subtracting the same sequence of values. It is
vital that a key used with this method should never be re-used, and competent
security tends to involve really careful attention to many details that do not belong
in this course18 .
Code the scheme described above. Print the first dozen output values from
it for your chosen key. You may like to check with a friend to see if their implementation generates the same sequence as yours when given the same key —
by the nature of this code there is not obviously going to be any other way of
characterising the correct output!
16
In the next section we will also see that it can be achieved by writing something like
(s[i] + k[i]) & 0xff.
17 “modulo”
18 But
means just the remaindering operation.
which will be covered later in the CST.
4.2. OPERATORS AND EXPRESSIONS
71
4.2 Operators and expressions
The examples shown already have included uses of the usual arithmetic operators,
both as used on integers and on floating point values. Now is the time to present
a systematic list of the operators that Java provides and the way they are used
in expressions. One of the critical things in any programming language is the
syntactic priority of operators. For instance in normal usage the expression a +
b × c must be read as if it had been a + (b × c) rather than as (a + b) × c. To stress
the importance of knowing which operators group most tightly I will list things
ordered by their syntactic precedence rather than by what they do. The simple
arithmetic cases will be listed but not discussed at any great length.
++, --: We have already see use of ++ as a postfix operator that increments
a variable. The full story is that the expression ++a has the side-effect of
increasing the value of the variable a by 1 and its overall value is that incremented value while a++ increments a but the value of the expression is
the original value of a. The use of -- is similar except that it subtracts one
rather than adds one to the variable mentioned. These operations can apply
to either integer or floating point variables;
+, - (unary): Unary + does not do anything but is there for completeness. Unary
- negates its (numeric) argument;
˜: The ˜ operator treats its integer operand as a binary number and negates each
bit in the representation. If you look back at the earlier table that illustrated
binary numbers you can check that ˜0 will have the same value as -1;
!: The ! operator may only be applied to a boolean operand, and it complements
the logical value concerned, so that !true is false;
(type): Casts are included here to show their precedence and to point out that
as far as syntax is concerned a cast acts just as a unary operator;
*, /, %: Multiplication, division and remaindering on any arithmetic values. The
odd case is the % operation when applied to floating point arguments. If
x % y is computed then Java finds an integer value q that is the quotient
x/y truncated towards zero, and then defines the remainder r by x = qy + r.
If integer and floating point values both appear in the same expression the
integers are promoted to the floating type before the arithmetic is performed.
Similarly if integers of different lengths are mixed or if floats and doubles
come together the arithmetic is performed in the wider of the two types;
+, -: Both integers and floating point can be added and subtracted much as one
might expect;
72
CHAPTER 4. BASIC USE OF JAVA
+ (string): If the + operator is used in a context where at least one argument is
a string then the other argument will be converted to a string (if necessary)
and the operation then denotes string concatenation. We have seen this used
as a way of forcing a conversion from a numeric type to a string ready for
printing. Note that the concatenation step will generally involve allocating
extra memory and copying data from each of the two original strings, so
it will tend to be much more expensive that the arithmetic uses of the +
operator.
<<, >>, >>>: Consider an integer as represented in binary. Then the << operator
shifts every bit left by a given number of places, filling in at the right hand
end with zeros. Thus the program fragment:
for (int i=0; i<32; i++)
System.out.printf("%d : %d%n", i, (1 << i));
will print out numbers each of which have representations that have a single
1 bit, with this bit in successive places. The result is a table of powers of 2,
except that the penultimate line of output will display as a negative integer
and the final one will be zero! There are two right-shift operators. The
usual one, >>, treats numbers as signed values. A signed value is treated
as if they were first converted to binary numbers with an unlimited number
of bits. For positive values this amounts to sticking an infinite run of 0
digits to the left while for negative ones it involved preceding the number
with lots more 1 digits. Next the value is shifted right, and finally the value
is truncated to its proper width. The effect is that positive integers get 0
bits moved into the vacated high order positions while negative ones get 1s.
When shifting an int the shift amount will be forced to lie in the range 0
to 31, while for type long it can only be from 0 to 63. The special right
shift written as >>> shifts right but always fills vacated bits with 0. It is very
useful when an integer is being thought of not as a numeric value but as a
collection of individual bits;
<, <=, >, >=: The usual arithmetic comparisons are available, and I have already remarked that there are a few delicacies with regard to floating point
comparisons and NaNs, infinities and signed zeros;
instanceof: This will be discussed later;
==, !=: Equality and inequality tests. For the primitive types they test to see if
things have the same value. For other types (arrays and the object-types that
are introduced later on) the test determines if the things compared are “the
same object”;
4.2. OPERATORS AND EXPRESSIONS
73
&: On integer operands the & operator forms a number whose binary value has a
1-bit only where both of the inputs have a 1. For positive values, for instance
a & 0xf and a % 16 will always yield the same result. Long tradition of
languages where the “and” operator is significantly faster then division and
remainder means that many old-fashioned programmers will make what is
now maybe excessive use of this idiom. The & operator can also be applied
to boolean operand, in which case is means just “and”;
ˆ: Exclusive or. See below to compare inclusive and exclusive or;
|: Inclusive or. Note that for integer values a&˜b | b&˜a == aˆb and the
same identity holds for boolean values except that ! has to be used for the
complement/negation operation rather than ˜. Here are the truth tables for
inclusive and exclusive or:
“|” 0 1
“ˆ” 0 1
0
0 1
0 0 1
1
1 1
1 1 0
&&: In a boolean expression such as A & B if the value of A was false there is
perhaps no need to evaluate B. The simple “and” notation does not take advantage of this, but the alternate form A && B does. Apart from efficiency
this can only make a difference if evaluating the sub-expression B would
have side-effects. In general I think it is probably good style to use &&
rather than just & whenever a boolean expression is being used to control an
if or similar statement, while & is probably nicer to use when calculations
are being performed on boolean variables;
||: This is the version of the “or” operator that avoids processing its right hand
operand in cases where the left hand one shows that the final value should
be true. Its use is entirely analogous to that of &&;
?: It is sometimes nice to embed a conditional value within an expression, and
Java lets you do that using the slightly odd-looking syntax a ? b : c. This
expects a to be of type boolean, while the other two operands can have
any type provided that they are compatible. If a is true the result is the first
of these expressions, otherwise it is the second. For instance the messylooking expression
(a==0 ? "zero" : "non-zero")
has the value "zero" if a is zero and "non-zero" otherwise. The phrase
(a || b) could be replaced by the equivalent form (a ? true : b),
while (a && b) has the same meaning as (a ? b : false);
CHAPTER 4. BASIC USE OF JAVA
74
=: Assignment in Java is just an operator. Thus you can assign to a variable
anywhere within an expression. The value of a sub-expression that is an
assignment is just the value assigned. Thus silly things like ((a=a+a) +
(b=b-1)) are good syntax if not good sense. A more benign use of the
fact that assignments are just expressions arises in the idiom (a = b = c
= d = 0). The assignment operator associates to the right so the example
means (a = (b = (c = (d = 0)))) and thus assigns zero to all of the
four variables named;
*=, /=, %=, +=, -=, <<=, >>=, >>>=, &=, ˆ=, |=: These operators combine
assignment with one of the other operators that have been listed earlier.
They provide a short-hand notation when the same thing would otherwise
appear to the left and immediately to the right in an assignment. For instance
a = a+3 can be shortened to a += 3 The abbreviation has slightly more
real value when the assignment concerned is to some variable with a name
much longer than just a, or especially if it is into an array element, since the
short form only has to evaluate the array subscript expression once. This
can lead to a difference in meaning in cases where evaluating the subscript
has a side-effect, as in the artificial fragment
int
int
int
for
[] a = ...;
[] b = ...;
p = 0, q = 1;
(int i=0; i<10; i++)
a[p++] += b[q++];
where index variables p and q step along through the arrays a and b and it
is important that they are each incremented just once each time around the
loop.
4.2.1 Exercises
Bitwise operations on integers
Investigate, either on paper or by writing test programs, each of the following
operations. Explain what they all mean, supposing that the variables used are all
of type int:
1. ˜a + 1;
2. a++ + ++b;
3. a & (-a);
4.2. OPERATORS AND EXPRESSIONS
75
4. a & ((1<<b)-1);
5. (a>>>i) | (a<<(32-i));
6. a + (a<<2);
7. (int)(byte)a;
8. (a & 0x80000000) != 0;
9. (a++ != b++) && (a++ == b++);
10. (--a != --b) | (--a == --b);
11. (a < 0 ? -a : a).
Counting bits in a word
Write a function that counts the number of non-zero bits in the binary representation of an integer. You can first do this using a test for each bit in the style
a & (1<<n) != 0. Next see if any of the examples in the previous exercise
give you a way to identify a single bit to subtract off and count. Consider also the
expression
(c[a & 0xff] + c[(a>>8) & 0xff] +
c[(a>>16) & 0xff] + c[(a>>24) & 0xff])
for some suitable array c.
What does this do?
This exercise and the few following it introduce a few fragments of amazingly
twisted and “tricky” code. Please do not view the inclusion of these programming
techniques here as an indication that you will be examined on them or that you
are being encouraged to use such obscure constructions in your own programs.
It is more that puzzling through these examples can refine your understanding of
the interactions between the Java arithmetic operations such as + and % and the
ones that work on the binary representations of numbers, ie & and >>. In a few
circumstances the ultra-cunning bit-bashing might save time in a really critical
part of some program and so could be really important, but almost always clarity
of exposition is even more important than speed. Certainly use of these tricks will
not make your programs any shorter, in that the bulk of the comments needed to
justify and explain them will greatly exceed the length of more straight-forward
code that has the same effect!
CHAPTER 4. BASIC USE OF JAVA
76
Start off with a any positive value. Execute the following19 and discuss what
value gets left in a at the end. Hint: look at the binary representation of a to start
with.
a
a
a
a
-= (a>>>1) & 03333333333;
-= (a>>>1) & 03333333333;
= ((a>>>3) + a) & 0707070707;
= a % 63;
And this. . .
a &= 0x3f;
a = ((a * 02020202) & 0104422010) % 255;
And this. . .
c = a & -a;
r = a + c;
a = (((r ˆ a) >>> 2) / c) | r;
[In each case it will help if you look at the numbers in binary.]
Integers used to represent fractions
Consider a as a value expressed in binary but now as a positive fractional value
in the range 0 to 1. This means that there will be an implicit binary point just to
its left. Then 0xffffffff will be just a tiny bit less than 1, 0x80000000 will
stand for 1/2 and 0x40000000 for 1/4. In terms of this representation interpret
the effect of executing the following four statements one after the other.
a
a
a
a
+=
+=
+=
+=
a
a
a
a
>>>
>>>
>>>
>>>
2;
4;
8;
16;
Division and Shifts
If a is a positive integer than a/2 and a>>1 give the same result. What is the
relationship between their values if a is negative. Carry on the analysis for division
by 4, 8, 16,. . . and the corresponding right shifts.
19 Discussions
with Alan Mycroft caused this and some of the other curious examples here to
re-surface. For a collection of real programming oddities including some of these try searching
for “Hakmem” on the World Wide Web or find “The Hacker’s Delight”[15] in a library
4.3. CONTROL STRUCTURES
77
Some Exclusive-Or operations
What is the final effect on a and b of the sequence
a ˆ= b;
b ˆ= a;
a ˆ= b;
Sieve for primes
Create an array of type boolean and length 1000. Set all elements to true to
start with. Then set items 0 and 1 to false
Repeat the following two steps
1. Find the first true item in the array. If there are none left then exit. Print
out the index of the value you have found, and call it p.
2. Set each item in the array that is at a position that is a multiple of p to be
false, for instance as in
for (i=p; i<1000; i+=p) map[i] = false;
The numbers you have printed should be the primes up to 1000.
If you wanted to find the primes up to several million (for instance to count
them rather than to tabulate them) it would make sense to make the array represent
just the odd numbers not all numbers. It might also save significant amounts of
space to represent the array as an array on int rather than boolean and pack
information about 32 odd-numbers into each int. You might note that some
programming languages can implement boolean arrays in that way without much
user intervention — Java does not.
4.3 Control structures
4.3.1 Exercises
Ambiguous If
Consider the sample code fragment
if (a == 0)
if (b == 0) System.out.println("Both 0");
else System.out.println("Some other case");
78
CHAPTER 4. BASIC USE OF JAVA
Figure 4.3: Keep control structures simple!
4.3. CONTROL STRUCTURES
79
and wonder exactly what happens if one or other of a or b is non-zero. Write
sample programs that test the actual behaviour of the real Java compiler either to
discover how Java resolves the near-ambiguity in syntax that this example represents.
Periodic Forests of Stunted Trees
The forms explained here were investigated by J C P Miller who was a lecturer
here in the Computer Laboratory. Their study leads on into that of error correcting
codes and so is perhaps less detached from the serious technical side of computer
science than one might think.
A root-line for a forest is a periodic binary sequence. Since it is hard to draw
things that repeat indefinitely it is useful to display such sequences by showing
one or two cycles and than agreeing that the ends of display should be treated as
being joined up to make a circle. Here is a sample root-line:
. X . . X X X . . X . . X X X . . X . . X X X .
A forest grows from a root-line by the simple rule that a branch grows when and
only when exactly one cell in the row beneath it is present (this is an exclusive OR
operation). I am drawing lines upwards20 following this rule from the root-line I
showed above yields
? . . . . . . . . . . . . . . . . . ?
? X X X X X X X X X X X X X X X X X X ?
? X . X . X . X . X . X . X . X . X . X ?
? X . . X X . . X X . . X X . . X X . . X ?
? . X X X . X X X . X X X . X X X . X X X . ?
? X X . X . . X . X X . X . . X . X X . X . . ?
? . X . . X X X . . X . . X X X . . X . . X X X ?
and the pattern seems to have died. In many cases after a number of rows the original row will re-appear21. Triangular clearings appear on the way. The challenge
is to understand when a pattern will die and when it will repeat, how the vertical
repetition period relates to the original horizontal one, and how large the largest
clearings will be. Write a Java application or Applet that takes a command-line
argument or otherwise accepts information from the user in the form of a string of
X and . characters. It should then display the generated forest.
20 If
you print things to System.out it would probably be easier to print with growth downwards. If you draw things to the screen in an Applet putting it either way up is easy. Observe that I
have drawn question marks to show where the pattern depends on data beyond the initial segment
of pattern. If you can allow for the (infinite) repetition of the base-line you do not need these.
21 You will see that I have only drawn a finite section of the infinite repeating base-line, and so
the forest is drawn as getting narrower as you grow it upwards. However it should be understood
to be of infinite width and so patters can re-occur exactly.
CHAPTER 4. BASIC USE OF JAVA
80
Life
The world consists of an infinite sheet of graph paper. Each
square may at any one time be either black or white. Every square has eight neighbours Every so often all squares
simultaneously follow the following rules:
3
4
5
2
6
1. A square that is black at present remains black if it
has two or three black neighbours. Otherwise it turns
white;
⊙
1
8
7
2. A white square becomes black if it has exactly three
black neighbours.
These rules define a behaviour which was invented by John Conway
(who at the time was in the Mathematics department here) and which
is known as Life. One starts off with a board that has a small number
of black seed points and display the position as the generations go
by. There are many astonishingly complicated things that can happen
and people have designed starting positions that illustrate them. The
challenge here is to make the computer run the rules and display the world-state.
A useful starting configuration to try has just five black cells arranged as at the
head of this paragraph. It explodes and seethes for quite a long while before the
situation stabilises. One thing to note is that all decisions about the next generation
are expected to be taken simultaneously, so any program that updates the world
incrementally is liable to get wrong answers. A further problem is that the the
ideal playing surface for Life is infinite, while computers tend not to be. Two
resolutions to this are usually considered. One places an immutable wall of white
cells as a border around the world so that all activity is contained within them.
The other scheme often used is to use a finite playing area but considers its left
hand column to be adjacent to its right most one and its top to be adjacent to its
bottom row. This amounts (depending on how you think of it) to playing Life on
a torus or to ensuring that all initial positions are replicated in a periodic manner
across an infinite plane.
The easiest program for this will set up two boolean arrays. The first holds the
current generation, while the second will be filled in with the next. My version
of a program that does this, complete with code to set up the initial pattern that I
have suggested and to draw the board positions in an applet window is around 75
lines long. I used a 200 by 200 board and kept the outermost rows and columns
permanently blank. That means that when accessing neighbours I can read from
them without going outside my array. Clearly the first exercise here is to reproduce
something like that.
4.3. CONTROL STRUCTURES
81
There are then three follow-up challenges. The first looks back to the optional
part of the binomial coefficient tickable exercise: can you get away with just one
boolean array rather than two, possibly keeping a boolean vector to store just one
row of backup information but mostly updating the world in place. To do so would
save around half of the memory that the simple program uses.
The second challenge observes that representing the
playing area as arrays of type
boolean is probably wasteful.
This would be a typical application where packing 32 cells
into an int and using lots of
bitwise and, or and shift operations to deal with them would
be common practice. It would
of course be possible to achieve
this by having nice abstract
procedures to reference the bit
Figure 4.4: Gosper’s Glider Gun.
at position (x, y) in an array
even though the array was being represented in a packed way. But it would perhaps give big speed savings to
look for ways to exploit the fact that bitwise operations on integers can handle 32
bits all at once and to try to use this to compute new values for several cells at the
same time.
Finally, and given that this program tends to run a little slowly, one looks at
where the time goes. Much of it will be wastage on parts of the board that are
totally white and hence where nothing is going to happen. Try to speed your code
up by avoiding as much of such wasted as is reasonable.
Eight Queens
Count the number of ways of placing eight queens on a chess board so so that
no pair are in the same row, column or diagonal as each other. This is a classical
puzzle to go in an introduction to programming and there are lots of clever tricks
that can be used. It is the sort of thing that most supervisors will have come across
before so I will not provide a fully worked through solution here, but I might
observe that the search might well be done by a recursive function that when
called at depth n will try to place a queen on row n of the chess-board.
82
CHAPTER 4. BASIC USE OF JAVA
Permutations
In Java arrays can be passed as arguments and newly created arrays can be returned as results. Write a function that accepts an array of strings as its argument,
and which hands back and array whose elements are themselves arrays of strings
giving all possible permutations of the input. For instance if I use curly brackets
to denote arrays here one might like {"a", "b", "c"} to turn into {{"a", "b",
"c"}, {"a", "c", "b"} {"b", "a", "c"}, {"b", "c", "a"}, {"c", "a", "b"},
{"c", "b", "a"}}.
As with several of the other Java exercises I might suggest that you design and
test an ML version first.
4.4 Control structures Part 2
There are two aspects of syntax that I will put off until a yet later section. One if
the syntax associated with the word class that we have seen wrapped around
every program we have written. The other is the matter of the “.” that appears between or possibly within so many names, eg System.out.println and
g.drawString. A few other bits of syntax will just not be covered in this first
course, although you may find traces of them in the grammar and discussion of
them in textbooks — and possibly also in next year’s “Concurrent Systems and
Applications” course.
But I will talk through each of the important components of the syntax and
give at least one illustration of each.
4.4.1 Expression Statements
Certain sorts of Java expression can be used as a statement — all that is necessary
is to stick a semicolon on the end of it. The cases permitted are where evaluating
the expression might have a side-effect. Thus an assignment expression, a function
call or a pre- or post-increment expression can be used. As an example, consider
the statement x++; which just increments x. An example such as 1+2+3; is not
considered valid: it would calculates the value 6 and then throws it away!
4.4.2 Blocks
Several statements can be placed one after the other to make a single large statement. Braces { . . . } are used around the statements to group them. In various
earlier languages the keywords begin and end were used instead of braces, but
Java prefers the version where you type in fewer key-strokes. If you see a block
4.4. CONTROL STRUCTURES PART 2
83
with semicolons in the semicolons are just parts of expression statements and
nothing special to do with the fact that there is a block there. Again some earlier
languages differed by using semicolons between statements in a block rather than
as termination of expression statements. Blocks can be nested any way you want.
You may insert extra braces to stress the grouping of any collection of statements
you feel deserves that, in much the same way that extra parentheses can always
be used to emphasis the grouping within expressions. I think there are enough
examples of blocks throughout these notes that I do not need to give a special one
here.
4.4.3 Null statements
If you insert a stray semicolon into a Java program it (mostly) does not matter
much, since a semicolon alone can be interpreted as an empty statement that does
nothing. The most striking example of the use of a null statement is in something
like
if (a > 7);
else System.out.println("Gotcha");
where the if needs a statement before its else part but no real action is needed.
If you really need such a place-holder I would suggest that the following is clearer
and flags your unusual intent more clearly.
if (a > 7) {/*NOTHING*/}
else System.out.println("Gotcha");
Better yet rearrange your code to make tests happen in a positive sense:
if (a <= 7) System.out.println("Gotcha");
4.4.4 if
It takes a little while to get used to the fact that the condition tested by if is
written in parentheses. Some people prefer a style where the statement after an
if is always written as a block, even if it is only a single statement, so that the
range that the if controls is made very explicit. This point of view has some sense
behind it, especially if the statement after the if is more than half a line long.
The control expression used by if must be of type boolean and so equality
tests are written as in a==0 and not a=022 . Using a single rather than double
equals sign is a common slip.
22 Which
would be an assignment and would have type inti.
CHAPTER 4. BASIC USE OF JAVA
84
4.4.5 while, continue and break
A while loop executes its command repeatedly for so long as the guarding expression remains true. Its syntax is very much like that of if. Within the iterated
command you can embed a statement “break;”, and execution of that will cause
a premature exit from the loop. The command continue;23 can be useful if the
iterated command is a long block, and it causes the loop to proceed at once to its
next cycle. Both break; and continue; are very convenient at times, but it is
often good style to avoid them when reasonably convenient so that the boolean
expression at the top of the while loop represents a total statement about the circumstances in which it will loop and when it will terminate. The following sample
shows a fairly typical while loop. Look back at your Discrete Mathematics notes
for explanation of why it computes a highest common factor and to give clues to
a reason for carrying out the extra computations. You may also like to code up an
extended Euclidean algorithm as a function that calls itself (say in ML rather than
Java) and observe that use of while loops does not always lead to the shortest or
most transparent code.
int a = 72, b = 30;
int u = 1, v = 0;
while (b != 0)
{
int q = a / b;
int r = a - q*b;
a = b;
b = r;
int t = u - q*v;
u = v;
v = t;
}
// Here a is the HCF. What are u, v?
Note that break can be used to exit other loops, and it is also used with
switch statements, which will be described soon.
4.4.6 do
Sometimes the most natural way to write a loop puts the test of a termination
condition at the end of the loop rather than at the start. This circumstance is
supported by the do statement, although I find it much less useful than while. In
fact I will often express
do
23 Observe
that the syntax for each of these command includes a semicolon, The identifier mentioned in the full grammar is something I will not discuss here.
4.4. CONTROL STRUCTURES PART 2
85
{
...
} while (xxx);
by writing it instead as
while (true)
{
...
if (!xxx) break;
}
since I think that do puts the details of what the loop is about rather too far down
the page. Anyway that also gave me a chance to include an example of a break
statement for you! The issues of programming style here could give rise to a
variety of discussions. A good policy is to try rather hard to make it very clear and
obvious just when each loop you write is going to terminate, and indeed to make
it clear (in comments as necessary) why you know it will eventually terminate.
4.4.7 for
Iteration with for has been seen in several examples. What is shown in the Java
syntax is that each of the three components within the parentheses and separated
by semicolons is optional. The most extreme case is when none are present:
for (;;) { ... } means just the same as while (true) { ... }.
In for (A;B;C) the expression A is an initializer evaluated just once at the
start of the loop. B is a boolean expression and is used just as in a while statement
to determine when to terminate. Finally C gets evaluated between each cycle of the
loop, and it often increments some variable. The idiom for (i=0;i<N;i++)
executes its command N times counting from 0 to N-1. The alternative way of
writing things is for (i=1;i<=N;i++). It loops the same number of times but
is maybe slightly less commonly used. Of course with the second version the
variable i starts at 1 not 0: this typically makes it less suitable for use as an array
subscript because in Java subscripts start at 0.
4.4.8 switch, case and default
There are occasions when one wants to dispatch to many different code fragments
based on the value of some expression. This could be achieved by writing a chain
of if .. else statements, but often switch provides a much neater way of
expressing things.
The construction starts with switch (Expression). The expression given
must be of type char, byte, short or int. Note that long is not allowed. The
86
CHAPTER 4. BASIC USE OF JAVA
switch-header is followed by a block enclosed in braces, and within this block
there can be special switch labels. The usual sort reads “case Constant:” and
control arrives just after the colon if the integer value of the switch expression
agrees with the constant after case. It is often useful to specify what action
should be taken if none of the cases that have explicit coverage happen, and for
this a label “default:” can be set. Case (and default) labels do not disturb the
usual sequential execution of statements, and so unless something special is done
after one case is processed control will proceed to the next one. This is usually not
what is wanted. A break; can be used to exit from the entire switch block. Many
programmers would count it is good style to put an explicit comment in whenever
a break is not being used, to show that its omission was deliberate and not an
accident.
If no explicit default label is given but a switch is executed in such a way that
none of the cases match it just acts as if there had been a default label just before
its final close brace.
It is generally a good thing to use switch whenever you have more than
three or four options to select between, in that it tends to be much clearer and
easier to understand than length strings of nested if statements. In the following
rather silly example it is imagined that the user has provided the function show
Observe that the case labels do not have to be in any special order, and that a
single statement can be attached to several labels.
switch ((int)n)
{
case 2: show("the only even ");
// drop through
case 3: case 5: case 7:
case 11: case 13: case 17:
show("prime\n");
break;
case 4: case 9:
case 16: show("square\n");
break;
case 8: show("cube\n");
break;
default:show("dull or too large\n");
// now just drop off the end
}
4.4. CONTROL STRUCTURES PART 2
87
4.4.9 return
When a function has done all it needs to it will want to return a result. This is
achieved using the return statement. Function definitions (see later) always indicate what type of result is required. They may have used the keyword void to
indicate that no result is needed. Such is the case with main. For void functions
one just writes “return;”, while in all other cases the syntax is “return Expression;”.
4.4.10 try, catch and throw, finally
Real programming languages need to be able to implement code that can recover
from errors and handle unusual cases tidily. The handling scheme in Java uses
the throw statement to raise exceptions. Throw statements specify an object
which should generally24 be of type Exception25 . The effect is that control
exits from the current block or procedure and any enclosing ones, all the way
until a suitable handler is found. If no such handler is present the computation
is terminated. The system has a number of built-in exceptions it will generate.
For instance an attempt to divide by the integer 0 raises an exception of type
ArithmeticExpression. Various functions that read from files can raise exceptions to indicate that the file did not exist, the current user did not have permission to read it or an attempt was made to read data beyond its end.
Handling exceptions involves prefixing a block with the word try and adding
on the end of it one or more clauses that describe what to do in unusual cases.
A clause that starts catch (Argument) is followed by another block which gets
obeyed if the system detects an exception whose type matches that declared for the
Argument. A single try may be followed by catch handlers for several different
types of exception.
try
{ z = 1/0; } // raises an exception!
catch (ArithmeticException e) { ... }
After all catch clauses you can put the keyword finally followed by another
block. The intent here is that this block will get executed come whatever, and it
will usually be used to tidy files or data-structures that the program might otherwise have left in a mess. A typical scheme to provide robust access to files would
go something like
<open the file for reading>
try
24 I
do not want to give the full and precise rules here!
be more precise of some type derived from Exception.
25 To
CHAPTER 4. BASIC USE OF JAVA
88
{
while (true) <read-more-from-file>
}
catch (<end-of-file-exception>)
{
// whole file read here. Good!
<success code>
}
finally
{
// must tidy up even if some failure
// other than end-of-file intruded
<close the file>
}
Later on I will give concrete examples that fill in the function calls and so on in
this framework.
Some programmers view catch and throw as neat and convenient language
features to be used wherever they fit. Certainly the file-handling example above
makes very good use of them. Others, and I tend to fit into this category, would
like to see them used rather sparingly in code since they can result in all sorts
of loops and functions terminating unexpectedly early and therefore undermine
attempts to make absolute statements about their end results.
4.4.11 assert
A statement of the form assertExpression; will evaluate the expression (which
really ought not to have any side-effects. If its value is false and if some magic
flag was supplied when the Java launcher was run then an exception is raised.
Assertions can have a second expression that can be used to give more details of
what you thought had gone wrong. It is proper style to include them at places in
your code where there is some reasonably cheap consistency check that you could
apply and when used well they are a huge aid to testing and debugging.
If you run your program normally the assertions will not be checked, and
furthermore having them in your source file will not hurt performance enough to
notice. If however you run the java command with the extra flag -ea the extra
checks will be done. Usage such as
java -ea:CheckThisClass SomeClass
will arrange to check just the assertions in the named class.
4.4.12 Variable declarations
Variable declarations can occur anywhere within a block. They are also allowed
in the first component of a for statement. The scope of a variable that is declared
4.4. CONTROL STRUCTURES PART 2
89
within a block runs from the declaration onwards until the end of the block. A
declaration made in a for statement has a scope that covers the remainder of the
for statement, including the end-test and increment expressions as well as the
iterated block. In fact the scope of a declaration appears to include the initialiser
for that variable, but if you try to use the variable there you should expect at least
a warning message. So things like
{
int x = x+5;
...
}
should not be attempted! A local consists of a type, then the name of the variable
being declared, and optionally one or more pairs of square brackets (to denote the
declaration of an array). Any initializer follows an “=” sign, and for arrays the
initializers are written in braces so that many individual values can be given so
as to fill in the whole array. A local variable declaration can be preceded by the
word final, and this marks the variable that is being declared as one that will not
subsequently change. A convention is that constants should be spelt entirely in
upper case, as have the examples PI and PLAIN that have been seen so far. Here
is an example:
final double E = 2.718281828459045235;
E = 1/E; // NOT valid because of "final"
4.4.13 Method definitions
A function definition starts with some optional qualifier words. The available
words are
public
abstract
protected
final
private
native
static
synchronized
and if present these can be written in any order. I will explain what they mean
later on. Next comes the type of result the function will return, which is either
an type or the special word void to indicate “no result”. Next is the name of
the function that is being declared, followed by a list of formal arguments (in
parentheses). A formal argument must be given a type, and may be preceded by
final if the body of the function will never update it. The grammar shown earlier
indicated that pairs of square brackets may be written after the formal parameter
list, but this should not be used in any new code26. If the execution of the function
can cause an exception to be raised and this exception is to be caught somewhere
then the fact must be mentioned by following the list of formal parameters by the
26 It
is a concession to some earlier versions of Java where it could be used for functions that
returned arrays.
CHAPTER 4. BASIC USE OF JAVA
90
keyword throws and then a list of exception types. Finally there is a block (ie
some statements within braces) that forms the body of the function that is being
defined.
For the moment you will still have to take the qualifiers public and static
on trust. They relate to the construction of the class that the whole file defines.
4.4.14 Exercises
Concerning 3n + 1
Take any number n. If it is even then halve it, while if it is odd replace it with
3n + 1. Repeat this process to see what happens in the long run. For various very
small integers you will find that you end up in a cycle 1 → 4 → 2 → 1 . . . but it is
not at first clear whether this is the ultimate fate when you start from an arbitrary
integer.
Write a program that generates the sequence starting from each integer from
1 to 1000. If the sequence ends at 1 record the number of steps it took to get
there. If you have taken over 10000 steps on some particular sequence then stop
and report just that value: after all maybe the sequence starting from that seed
goes on for ever, either by diverging to infinity or by finding a cycle different from
the one that includes 1. If on the way you generate an odd number larger than
(Integer.MAX VALUE-1)/3 you should also stop the calculation there since
otherwise you would suffer from integer overflow and subsequent work would be
nonsense. The constant Integer.MAX VALUE is another Java built-in constant
useful in cases such as this.
Arrange that you only print anything when a new record is broken for the
length of a sequence or when you would reach integer overflow. For each recordbreaker display the seed, the number of steps taken before 1 is reached (or the fact
that an overflow occurs) and the largest value in the sequence concerned.
Tickable Exercise 4
As you start this exercise note that ticks 1, 2, 3 and 4 are probably fairly easy. Tick
5 is going to be a somewhat larger piece of work so as soon as you have finished
this one you might like to look ahead and get started on it!
In ML a function called quicksort could be defined as
fun select ff [] = []
| select ff (a :: b) =
if (ff a) then a :: select ff b
else select ff b;
4.4. CONTROL STRUCTURES PART 2
91
fun quicksort [] = []
| quicksort (a :: b) =
quicksort (select (fn p => p < a) b) @
[a] @
quicksort (select (fn p => p >= a) b);
The idea is to use the first element of the input list as a pivot. One then selects out
first all the remaining values that are less than this pivot, and all the values that are
at least as large. Recursive calls sort the two sub-lists thus generated, and a final
and completely sorted list is obtained by concatenating the various parts that have
been collected.
The ML version is very elegant and shows some of the important ideas behind
the Quicksort algorithm. However it misses out several other things that are important in the real Quicksort method, mostly issues concerning use of memory.
In this exercise you are to implement a version of Quicksort in Java. You should
write a procedure with signature27
void quickSortInner(int [] v, int i, int j)
which will sort that part of the array v that has subscripts from i to j. It will
be up to you to decide if these limits are inclusive or exclusive. The procedure
should work by first seeing if the sub-array it has to work on is empty. If so it can
return without doing anything! Otherwise it should take the first (active) element
of the array as a pivot and rearrange28 the remaining items so that the array gets
partitioned at a point k such that the pivot has been moved to position k, all items
to the left are smaller than the pivot and all items to the right are at least as large
as it. It should do this re-arrangement without using more than a few extra simple
variables: ie it is not acceptable to create a whole fresh array and copy material
via it. quickSortInner can then call itself recursively in a way suggested by
the ML code to complete the sorting process.
You should also define a function called just quickSort that takes only one
argument — the array to be sorted. Remember that the .length selector can tell
you how large the array is.
To show that your code works you should demonstrate the following tests:
1. Create an array of length 10 and show the effect of sorting it when its initial
contents are (a) the numbers 1 to 10 starting of in the right order to begin
with, (b) 1 to 10 in exactly the opposite order to begin with, (c) ten numbers generated by nextInt() from the random number package (d) ten
numbers all of which are zero;
27 The
signature of a function is just the specification of the types of its arguments and result.
the National Flag exercise.
28 Remember
CHAPTER 4. BASIC USE OF JAVA
92
2. Measure the time taken to sort various length vectors of random data where
you should use lengths 16, 32, 64, . . . up until the sorting run takes several
seconds. For each test compute the quotient of the time taken and the value
N log(N) where N is the number of items being sorted.
Optional part for those who are keen: Read the Java documentation for the
Array.sort(int []) method that Java 2 provides. Write code to time it and
compare the results with the code you wrote yourself. When measuring times
work with arrays long enough that each test takes several seconds. Observe
that the fact that the Java libraries provide you with sorting methods (see also
Collections.sort) means that most Java users will never need to implement
their own Quicksort: you are doing it here as an exercise and because it is good for
Computer Scientists to understand what goes on inside libraries, since next time
around it may be their job to implement libraries for some new language.
As a further optional extension to this exercise consider the following and
adjust your code accordingly, then repeat all your tests:
1. The ML quicksort partitions items by comparing them with the value that
happened to be first in the list. In the plausible cases where the original
data is already in ascending or descending order this leads to excessive cost.
Selecting as the “pivot” the median of the first, last and middle element from
the array being sorted29 does better;
2. It is probably best to stop quickSort from recursing once it gets down to
sub-arrays of length 3 or 4. The end result is that it almost sorts the array,
but a final pass of bubble-sort can finish off the job nice and fast. Is this
born out in your code?
3. The partitioning code here can be delicate! Unless you are careful it can
escape beyond the bounds of the array, or it can get muddled about whether
the two final values in the middle of the array need exchanging or not. Simple implementations can be made safe by making all the end-conditions in
your loops composite ones rather like
while (k>=i && v[k] > pivot) ...
while if we could get away with it it should be faster to go something more
like
while (v[k] > pivot) ...
29 Always supposing
there are at least 3 items in the list.
4.4. CONTROL STRUCTURES PART 2
93
Investigate how well you can trim down your inner loops while retaining
code that always works! The Part IB course on Data Structures and Algorithms and the textbook by Cormen et al[9] are where this level of detailed
study really belongs!
(End of tickable exercise)
Highest Common Factors
Implement code to compute Highest Common Factors using the Euclidean Algorithm. Extend it to use the extended algorithms that at the end will allow you to
solve equations of the form
Au + Bv = 1
Tickable Exercise 5
The work called for here will be done in sections, and it is expected that while
working towards the tick you will be able to design, code and test each section
before moving on to the next. The idea involves creating a package of routines
that can compute with (univariate) polynomials. For the purposes of this exercise
a polynomial
(a0 + a1x + a2 x2 + . . . + anxn )/b
will be represented as an instance of a the class:
class Poly
{
private String variableName;
private long [] coeffs;
private long denominator;
... constructors and methods as needed
}
where variableName holds “x”, the array called coeffs stores the coefficients
a0 to an and the long denominator holds the value shown as b above. Because
Java lets you enquire as to the length of an array it is not necessary to store the
degree n explicitly. In this representation common factors should be cancelled
out between numerator and denominator, and the highest degree coefficient an
should never be zero. In this exercise all polynomials will be in terms of the
same variable, x, so the variableName should always be set to "x" and it will
not play much of a part in any of the calculations! Step by step carry out the
following tasks, testing what you have done as best you can as you go:
CHAPTER 4. BASIC USE OF JAVA
94
Create simple polynomials: Write functions that can create the “polynomial”
that represents just a given integer, a given fraction and the simple polynomial “x”;
Debug-quality printing: Write code that takes a polynomial and displays its coefficients. For this part of the exercise it is not at all important that the
display format you design be tidy or that it respects line-lengths. So for
instance you may generate output such as
(1*xˆ0 + 0*xˆ1 + -3*xˆ2)/2
with various unnecessary symbols in there. The object is to be able to see
your polynomials clearly enough that you can test and demonstrate what
comes next!
Special-case multiplication: Write code to multiply a polynomial by an integer,
to divide it by an integer, and to multiply it by x. Note that in the first
two cases you will need to do calculations (involving greatest common divisors30) to reduce the coefficients to lowest terms. In the latter case the
result will be of one higher degree than the input and so will be represented
with a coefficient vector one item longer. These routines should not alter
their input, but should create new polynomial data to represent the results;
Addition and Subtraction: Take two polynomials and create another that represents their sum (or difference). This involves more fun with ensuring that
the result is over a common denominator, and subtracting two polynomials can lead to a result of lower degree if the leading terms cancel (as can
adding if the leading terms start off as similar but with opposite signs);
Multiply: If you have one polynominal of degree m and one of degree n then
their product is of degree m + n. Write code that computes it;
Differentiate: in fact differentiation of a polynomial by its variable is rather an
easy operation (and so would be integration, which you would need in an
optional extra to this exercise). If the polynomial contains an original terms
ai xi then the derivative contains just (iai )xi−1 ;
Proof of pudding part 1: Let P0 = 1, P1 = x and from there on define a sequence
using the recurrence relationship
Pn = ((2n − 1)xPn−1 − (n − 1)Pn−2 )/n
30 Otherwise
known as highest common factor.
4.4. CONTROL STRUCTURES PART 2
95
Using your polynomial manipulation program calculate and tabulate the values up to (and including) P12 ;
Proof of pudding part 2: Now instead define
Pn =
1 dn 2
(x − 1)n
2nn! dxn
(This is known as Rodrigues formula, in case you wondered, and the polynomials you are computing are Legendre Polynomials)
Using this recipe compute and display values up to P12 . The two sequences
of polynomials you have computed ought to match!
Testing to destruction: extend your tables until the values computed by the two
recipies for Pn are incorrect because of some internal integer overflow, and
report where your program first displays a result that is certainly incorrect;
Optional extra (a): Write code that evaluates a polynomial at an integer value of
its variable, ie at x = n. Write code that computes the (indefinite) integral of
a polynomial with respect to its variable. Combine these two to allow you
to evaluate definite integrals. Display a table showing values of
Z 1
x=−1
Pi (x)Pj (x)
for i and j running from 0 to 5 (say);
Optional extra (b): Let y be one of the polymonials (Pn ) that you have just computed. Evaluate
(1 − x2)y′′ − 2xy′ + n(n + 1)y
Tabulate this for various small values of n.
Note: the examples worked with here are Legendre polynomials, and provide an
example taken from Sturm-Liouville theory. Optional extra (a) shows that they are
orthogonal over the range from −1 to +1 and this in fact makes them useful for
producing certain sorts of good numerical approximations to functions. Optional
extra (b) is showing you that these polynomials are solutions of a differential
equation: many other interesting sequences of functions satisfy recurrence formulae, have orthogonality properties and are solutions to differential equations!
Abramowitz and Stegun’s book of tables[1] is probably the easiest place to suggest you look to find out more.
(End of tickable exercise)
CHAPTER 4. BASIC USE OF JAVA
96
Pollard Rho integer factorisation
In previous years this was Tickable exercise 5. There are in fact a few delicacies
with regard to integer overflow (which do not greatly damage it as an exercise but
which could raise questions about it). You may still like to try it!
The explanation of this exercise is quite long, and it maybe looks messy, but I
can assure you that the code that has to be written is tolerably short and managable
once you have sorted out what needs doing.
Randomised factorisation: Implement the following algorithm that (possibly)
finds a factor of an integer that it is given:
A single trial that looks for a factor of N is performed by selecting a random
positive number R and computing S = R % N. This is a number between 0
and N − 1. If the number is 0 deem this trial a failure. Next compute the
highest common factor of S and N. If this is 1 then again the trial is deemed
a failure. However if the HCF is not 1 then it is a factor of N and because it
is also a factor of S it must be less than N. This counts as success!
The complete factorisation algorithm
works by running a number of trials.
√
If for a number N the first N trials all fail then we will pretend that N
is prime. Otherwise a factor of it has been found, and dividing this into N
gives us its co-factor. Smaller factors of each of these can then be sought
using the same technique.
Use this procedure to try to factorise the numbers 2i − 1 for i from 2 upwards, stopping when your program starts to take more than a second or so
to run.
The Birthday problem: Suppose we have a sequence of numbers all of which
are less than N, and these values are generated in some way such that each
number xn is some fixed function of xn−1. A concrete example would be if
xn = (x2n−1 − 1) % N. For most N and for x0 = 2 this sequence31 is in fact
pretty unpredictable.
Any such sequence must eventually repeat a value, and once it has it necessarily continues in a loop. If consecutive values behave well enough as if
they are random up to this point then the expected length of the sequence
31 There
is a significant delicacy here: when you compute x2n−1 its value can be almost as large
as
even though the remaindering will rapidly bring it down to a smaller range. This can lead
to integer overflow and a particularly un-wanted effect is that a value you generate may end up
unexpectedly negative (when N 2 is outside the valid range of integers). I suggest you mostly
ignore this here (!) and at most take an absolute value to ensure that the sequences you generate
consist of positive numbers. Also there is not much special about starting with x0 = 2 and other
randomish starting values might work just as well.
N2
4.4. CONTROL STRUCTURES PART 2
97
before a repeat is related to the problem of how many people you have to
have in a room before you should expect to find that two of them share a
birthday. In this case the room is on a planet in a galaxy far far away, where
the length of the year is N, and the statistics suggest that we need around
√
N of our aliens.
For this exercise you are to imagine one algorithm that detects a cycle and
implement a second and much better one.
The algorithm you just have to imagine guarantees to find a cycle as soon
as it arises. It allocates a big array and stores values in this array as they are
generated. As each one is generated it also checks through the ones already
seen to see if the new value has occurred before, and if so declare the loop
detected. This method is easy to visualise but it needs an array as long as
the longest potential loop, and the search means that before finding a loop
at step n it has done about n2 /2 comparisons with old values. This is slow.
The second method, which you should implement, records the value of xi
each time i reaches the next power of 2 and compares newly generated values against this one stored value. It argues that if there is a loop then eventually the loop will be totally traversed between consecutive powers of 2,
and thus will be detected. Furthermore this will be at worst a factor of two
beyond the place where the first repeat happened.
Having coded the second loop-detection algorithms try it on sequences generated by xn = x2n−1 − 1 mod N for various N and verify that for a reason√
able proportion of values of N and x0 a loop is found after very roughly N
steps.
Pollard Rho: This builds on the previous two parts, so please do not start it until
you have completed them. But then re-work the loop detection code so that
instead of comparing each new xn with a saved value x2k using an equality
test compute the HCF of N and xn − x2k . Stop if this is not 1, ie if a factor
of N has been found. In the case when xn = x2k the method has failed: you
may either give up in that case or try starting again with a different value
for x0 .
Implement this using the Java long type. If the number N is composite
it is probable (although
not guaranteed) that this will find a factor of N
√
4
within around N trials, and will thus be able to find a factor any Java
long value quite rapidly. Of course if N starts off as a prime this scheme
will never manage to find a factor of it! To test this you should probably
create numbers that are known to be composite by multiplying together two
int-sized values.
98
CHAPTER 4. BASIC USE OF JAVA
Optional: The scheme above does not of itself find a complete decomposition
of an integer into prime factors — it just splits composite numbers into
two. A complete fatorisation method needs to extend it with first a filter so
that numbers that are prime are not attacked, and secondly with recursive
calls that try to factor the two numbers that Pollard Rho split our number into. Investigate the Java BigInteger class that provides arithmetic
on long integers and which also provides a test for (probable) primality.
Re-implement your code to use BigInteger rather than long and to use
isProbablyPrime to avoid trying to factor when it is futile. Thus produce
code that can produce complete factorisations of reasonably large numbers.
How many digits long a number can you factorise in say 20 seconds?
Some of you no doubt consider yourselves to be Java experts! You may like
to arrange that the calculation x2n−1 − 1 mod N is computed exactly even
when N is almost as large as a Java long can be, and that overflow does
not interfere. An easy way to do this is to use the Java library big integer
support, but what I would prefer here would be code expressed entirely in
terms of use of long arithmetic.
4.5 Java classes and packages
What has been described thus far should provide a foundation for understanding
the small-scale structure of Java programs. If you have understood it you are
equipped to write programs that have up to (say) half a dozen sub-functions and
that are limited to living in a single source file. So far the data that Java can work
with has been limited to the primitive types int and so on, together with arrays
of them. The time has now come to discuss the Java idea of a class. This is used
both to support the construction of user-defined data-structures and to impose an
order on programs that are large enough that they should properly be spread across
several source files. A discussion of Java classes will include an explanation of
what all the “.” characters are doing in the sample programs seen so far. All of
this counts as “Object Oriented Programming”.
One of the aspects of programming language design that has proved to be
especially important is that control of the visibility of names. This whole issue
tends to look rather frivolous — a distraction — while your programs are only a
page or two long but it makes a critical difference to big (and perhaps especially
collaborative) projects. There are several interlocking reasons to want to keep
name-spaces under control. One as so that a large chunk of code can be given
a cleanly defined interface consisting of the functionality that it makes visible to
the outside world. Everything not so exported is then deemed private to the group
who maintain that body of code, and they may change internal parts of their design
4.5. JAVA CLASSES AND PACKAGES
Figure 4.5: Classes and Packages make Java “modern”.
99
100
CHAPTER 4. BASIC USE OF JAVA
with complete confidence that this can not hurt anybody else.
A second reason for keeping name-spaces well controlled is so that different
parts of a large program are free to re-use the most obvious names for their functions and variables, secure that this can not introduce unexpected clashes.
Related to both of these is the fact that when trying to understand code limits
on the visibility of names can allow you to concentrate on just the range in the
code where something is relevant.
Java controls access to names at three levels. At the finest grain it has scope
rules that are much like those of most other programming languages. If a local
variable is declared within a block that variable can only be referenced using code
textually within that block. Java understands the idea that a re-declaration of a
variable in an inner block would create a different variable with the same name,
and that within the inner block the new variable would shadow the old one, as in
int func(int a)
{
{
int a = 4, b = 5;
// ????
for (int a=0; a<10; a++) b++; // ????
System.out.printf("%d %d%n", a, b);
}
return a;
}
and it views this as something that could be codified and that to a computer has
a totally logical interpretation. But that it is a potential cause of real confusion to
human programmers so it should be prohibited! Thus the above example will be
rejected by the Java compiler and all the interesting computer-science discussion
of exact rules about scope can be set aside. You may like to note, however, the
variables do not clash in any way if their scopes do not overlap, so the following
is valid:
int func(int a)
{
{
int i = 4;
for (int j=0; j<a; j++) i++;
System.out.println(a + " " + i);
}
for (int i=0; i<10; i++) a *= 2;
for (int i=0; i<a; i++) a--;
return a;
}
The scopes associated with each declaration of i are disjoint.
The other two aspects of Java name-space control are more interesting. The
important words used here are class and package.
4.5. JAVA CLASSES AND PACKAGES
101
All names of Java variables and procedures32 live in some class. In general
you have to gain access to the class before you can use its members33. A member
of the current class can be referred to just by giving its unqualified name, but in
other cases you need to have access to an object of the required class and refer to
the member using a dot “.” as a selector on it. This is what was happening in cases
such as g.drawLine where g was a variable of type Graphics and drawLine
was a member of that class. When a class is defined the user can arrange which
of its members can be referenced by other classes in this manner, so that internal
details of the class can not even be accessed using this sort of explicit naming.
The word public flags a component of a class that should be universally visible
while private marks one that should not.
Classes thus contribute in two ways to the avoidance of confusion over names.
Firstly they mean that most references to things outside the current class will
include a dot selector that indicates fairly explicitly what context the name is to be
taken from, and secondly they can arrange that some names are kept totally local
to the class within which they are used and can never be accessed from anywhere
else. It is perhaps worth reminding you at this stage of the qualifier final that
can turn a variable declaration into the definition of a constant. There are further
refinements in the control of name visibility and use that Java provides, and the
keywords protected, abstract and static relate to some of them: these will
be discussed later on.
Classes themselves have names, and so a scheme is needed to structure the
name-space that they live in. A collection of classes can be placed in a “package”.
When classes are declared only those that have been given the public attribute34
are visible outside the package. Furthermore since the idea is that any other Java
code35 can access the public classes of a package there is a somewhat curious
linkage between package names and the filing system on your computer. This
linkage is mediated by a thing called the “class path” which can list the places
that Java should search to find the compiled code if you refer to a class defined
in some package. You can expect that any reasonable default Java setup will have
your class path set up for you already so that you can access all of the standard
Java libraries and so that code in the current directory can be used. The full names
of classes generally contain dots. Various names starting with the component
java are reserved for the system, and ones starting with sun are for use by Sun
32 From
now on I will increasingly move towards the Java notation and call these “methods”.
will see later that in some cases, when the name has been declared as static, one
can refer to the item via the name of its containing class. But in the more general case it will be
necessary to have an instance of the class and access the item via that.
34 Making a class public is a similar idea to making a member of that class public, but of
course we are talking now about a different level in the structure of a program.
35 Ideally anywhere in the world!
33 We
CHAPTER 4. BASIC USE OF JAVA
102
Computers, who designed Java. The various further parts to package names are
intended to group packages into hierarchies. For instance every package whose
name starts with java.awt is to do with the Java Abstract Window Toolkit, which
is the part of Java that provides facilities to pop up windows on your display. The
package java.awt.event is the sub-part of this that contains classes relating
to events — we have seen an example where these could be caused by the user
clicking with the mouse but there are others. The Java documentation contains a
list of all the predefined packages that are part of the Java core, and then lets you
browse the complete set of classes defined in each. Each class of course provides
a number of variables and methods: the number of standard library methods is
huge!
Specifying full names in the package hierarchy could become very tedious, so
Java provides a user-configurable way of setting up shorthand forms of reference.
Recall that various sample programs we have seen began with a collection of
import statements
import javax.swing.*;
import java.awt.*;
import java.awt.event.*;
This adjusted Java’s name resolution scheme so that the class MouseEvent (say)
could be referred to by that short name. Without the import it would still have
been possible to write the same program but it would have been necessary to use
a fully spelt-out “java.awt.event.MouseEvent” to name the class, and that
would involve knowing exactly which of the standard parts of the Java library
MouseEvent belongs in. The “*” in the import statements show tells Java to
support short names for all the classes in the packages named. It is also possible
to put a single class name in an import statement. This could be useful if only
one member of a package was to be used and you did not want to risk confusion
with other names from that package. Some Java programmers take the view that
import with a “*” introduces risk of giving them access to classes other than ones
the know about, and so always spell their imports out fully, despite that being a
little more verbose. Note that the syntax of Java only allows the wild-card “*”
to be placed right at the end of an import where it means “all the classes in this
package”.
If you issue import statements that attach to two or more packages that define
identically named classes then Java will refuse to get muddled: it just insists that
you use fully spelt-out names for the classes that could otherwise be ambiguously
resolved. This is probably safer than a scheme where the first (or possibly the
last?) import statement took precedence.
Java comes with around 200 and huge numbers of pre-defined classes, and so
getting to know them all is a big job. You are not expected to do so especially
4.5. JAVA CLASSES AND PACKAGES
103
as future Java releases will add yet more, and it is probable that when you work
on any big Java project you may find yourself using substantial third-party class
libraries. But it does make a lot of sense for you to have a good overview of what
is available to you so that when relevant you can use existing well-tested library
code rather than starting to write something of your won.
As well as being a way of organising the name-space all Java classes count
as data-types. When something’s type is a class it is usual to refer to the thing
as an “object”. Thought of in terms of objects, a class defines a data structure
that contains fields that are the variables defined in it and happens also to be able
to contain definitions of functions that will access these fields. If the variables
were declared public then any code anywhere can access them and so there do not
really have to be any (explicit) methods defined within the class. There will in
fact always be a few implicitly defined ones that are to do with the creation and
deletion of objects.
Taking a minimal approach36 to class definition I can now set up a definition
that would let me represent binary trees where each node in the tree contains an
integer:
// Compare the ML version, which would be
//
datatype Tree =
//
nullTree |
//
makeTree of int*Tree*Tree;
// or some such.
class BinaryTreeOfIntegers
{
public int value;
public BinaryTreeOfIntegers left;
public BinaryTreeOfIntegers right;
}
Comparison with the ML version reminds us that it is important to be able to have
some way of telling when the left and right children of such a tree do not really
exist. In ML that was achieved with an explicit alternative constructor, which I
called nullTree. In Java any variable which has a class37 as its type can either
hold a proper instance of that class (ie an object) or it can hold the special value
null. This value is provided as a keyword in Java. Ie the word null is hardwired into the Java language and not just some curious pre-defined variable. It
also has the odd property that the same value may be used with any sort of class
36 The
class I define here would work, but it misses out on exploiting a lot of structuring and
security features that classes can provide, and so is just a minimal start.
37 Or array.
CHAPTER 4. BASIC USE OF JAVA
104
or array variable to set the variable to a state where it “does not hold an any object
at all”.
Once a class has been defined it will be useful to declare variables using it and
create objects to go in them. Here I will create a rather small tree using the above
class definition:
...
BinaryTreeOfIntegers a1, a2, a3;
// a1, a2, a3 are all un-initialised here, and
// Java complains if you try to compile a program
// that relies on the values of variables that
// might not have been given a value.
a1 = new BinaryTreeOfIntegers();
a2 = new BinaryTreeOfIntegers();
a3 = new BinaryTreeOfIntegers();
a1.value = 1;
a2.value = 2;
a3.value = 3;
a1.left = a2;
a1.right = a3;
// the next 2 lines are not needed in that
// null is the default value given to a field
// that would hold an object.
a2.left = a2.right = null;
a3.left = a3.right = null;
...
Note (but do not worry about, for now) the parentheses after the class name following new. And also observe how dreadfully clumsy all this is.
Note that Java provides default initialisers for instance variable in classes and
elements in arrays, but not for local variables within methods. The default values
used are zero for numeric fields, false for booleans, ‘\0’ for characters and null
for all references.
Anybody who is a C or C++ programmer is liable to have a question to ask at
this stage, but those who have mostly seen ML should see all this as reasonable.
You can also see “.” being used as a selector to access the components of a class
object. The C programmers can read my footnote38! Java objects are created in
much the same way as Java arrays are, using new, and there is no need to take
38 In
C or C++ one would distinguish rather carefully between a structure and a pointer to the
structure. And in C terms all Java class variables hold pointers. However in Java it is not really
useful to think this way since all Java operations have been designed to prevent any explicit tricks
involving pointers. Please try to think of Java objects as more in the style of ML data. In C
you the explicit visibility of the difference between a structure that is directly at hand and one
that is referred to via a pointer leads to a distinction between the use of “.” and “->” to access
4.5. JAVA CLASSES AND PACKAGES
105
any special action when you have finished with one. The Java run-time system is
expected to tidy up memory for you. However grossly excessive object creation
can either consume time or utterly run you out of memory. The first loop in the
following code does not do anything very useful with the objects it creates, and it
discards them all rather rapidly. It may be a bit inefficient. The second loop creates
a million objects and chains them all together so that none of space concerned can
be recycled. At one million you may get away with this, but if you tried to do this
a few hundred times more your computer’s memory would not be able to keep up
with the demands of the program and an exception would be raised to report this
fact.
for (int i=0; i<1000000; i++)
{
BinaryTreeOfIntegers x =
new BinaryTreeOfIntegers();
x.value = i;
// x is implicitly discarded here
}
BinaryTreeOfIntegers w;
for (int i=0; i<1000000; i++)
{
BinaryTreeOfIntegers x =
new BinaryTreeOfIntegers();
x.right = w; // chain on to w
w = x;
}
There are very few cases in Java where it would be considered good style to
define a class that only had variables defined within it39. Mostly an attempt will
be made to collect almost all of the methods that work with the class as part of
it. Very frequently the variables in the class can then be made private, and
the public methods provide a clean and abstract interface to everything. There
is something of a convention about providing and naming methods to access the
data stored in an instance of a class: methods that update variables have names
starting with set, ones that retrieve boolean values start is while others that
retrieve information start with get. Here is the previous example expanded to
follow these conventions, and adjusted so that the case of boolean variables can
be illustrated:
components. Again Java does not need this and so only has one notation, even though in some
sense it uses dot where a C programmer would naturally reach for an arrow.
39 The most plausible good case I can think of is when all the variables are marked as final
so they are constants and the class is just used to encapsulate the name-space within which these
constants are defined.
CHAPTER 4. BASIC USE OF JAVA
106
// Compare the previous Java version where
// the variables were public but there were
// no methods.
class BinaryTreeOfBools
{
private boolean value;
private BinaryTreeOfBools left;
private BinaryTreeOfBools right;
public void setValue(boolean n) { value = n; }
public void setLeft(BinaryTreeOfBools t)
{ left = t; }
public void setRight(BinaryTreeOfBools t)
{ right = t; }
public boolean isValueTrue()
{ return (value==true); }
public BinaryTreeOfBools getLeft()
{ return left; }
public BinaryTreeOfBools getRight()
{ return right; }
}
For small classes this just adds way too much extra verbiage and feels silly. However for a large and compilicated class with many other methods having a regular
and predictable naming can be a real help. It also provides a way that you can
give read-only access to some variables or you can check the sanity of values
to be assigned to others, ending up with much finer-grained control over access
than even use of the public and private qualifiers give you. The term “bean”
is sometimes used for Java classes that follow this set of conventions, and some
programming tools exploit it. Because it makes small programs so much bulkier
I will not use this style in every example in these notes, but you can notice that
many of the Java library classes clearly have and you might think about it again
when you move on to writing large classes for yourself.
Here is a sample Java class that might be useful within other programs and that
illustrate methods that actually do something useful. It is a start at code that will
enable Java code to work with complex numbers. An odd-looking programming
style that it illustrates is one where to combine two complex numbers, say a and b,
one will call a method associated with one of them, passing the other as argument.
Thus the sum of the two values will be requested as a.plus(b). It is not possible
(in Java) to redefine or extend the basic “+” operator to make it “add” objects
from some new user-defined class, hence use of a method name such as add is
4.5. JAVA CLASSES AND PACKAGES
107
necessary here40.
public class Complex
{
private double x, y;
// define setX, setY, getX, getY here if you want.
public Complex(double realPart, double imagPart)
{
x = realPart;
y = imagPart;
}
public double modulus()
{
return Math.sqrt(x*x+y*y);
}
public Complex plus(Complex a)
{
return new Complex(x + a.x, y + a.y);
}
public Complex times(Complex a)
{
return new Complex(x*a.x - y*a.y,
x*a.y + y*a.x);
}
}
This would be placed in a file Complex.java and compiled using javac in the
usual way to make a file Complex.class. Because I have not put in a package
statement this class will live in a default package, and when other Java programs
run and they want a class called Complex they might manage to find this one if
its class file is still in the current directory.
The Complex class illustrates one new concept. Observe the method definition that uses the name of the class as its own name and which does not specify a
separate return type:
public Complex(double realPart, double imagPart)
{
x = realPart;
y = imagPart;
}
It has no return statement in it. A method whose name matches that of the class is
a constructor and you will typically use it with new to create fresh instances of the
class thing concerned. If you do not specify an explicit constructor function then
a default one is supplied — it has no arguments and does not leaves all variables
in their default state. It is valid to have several constructors provided that the types
40 In
contrast the language C++ does allow you to extend the meaning of all the operators that
are denoted by punctuation marks. Many people believe the conciseness and elegance that can be
achieved that way is more then balanced out by the potential for severe confusion.
CHAPTER 4. BASIC USE OF JAVA
108
of their arguments are different. Observe here how the methods that are members
of the class all have access to the private variables, but no code outside the class
will have.
Sometimes when referencing a variable it is useful to stress that you are talking
about one in the current instance. The keyword this always refers to the object
from which you invoked a method, and so the constructor and the plus methods
above could have been written out in a way that some would consider clearer:
public Complex(double x, double y)
{
this.x = x;
this.y = y;
}
public Complex plus(Complex a)
{
return new Complex(this.x + a.x, this.y + a.y);
}
Explicit use of this can be used to avoid mixups if the name of a formal parameter for a method matches the name of a variable in the class. Consider the
following and the muddle that would arise without the use of this, but also note
how much nicer it is to select names that avoid any hint of a clash.
public Complex plus(Complex x)
{
return new Complex(this.x + x.x, this.y + x.y);
}
4.5.1 Exercises
Complete the Complex class
The class as shown here does not support division, and does not have an equality test. If you define a method called toString() in it then will be called to
“print” the number when you use “+” to concatenate it with a string. Finish off
the Complex class adding in these and whatever other facilities you feel will be
generally useful.
Polar Complex Numbers
The Complex represents complex numbers in Cartesian form, ie as x + iy. But
the internal variables x and y that it uses are both private so nobody outside
the class can tell this! An alternative representation of complex numbers would
store a number as a pair (r, θ) where the complex value concerned had modulus r
and argument θ. In other words one would have z = reiθ . At the cost of computing a few arc-tangents and the like it is possible to create a re-worked complex
4.5. JAVA CLASSES AND PACKAGES
109
class that has exactly the same external behaviour as the original one but which
stores internal values in polar form. The constructor function and addition become
messier, multiplication becomes easier and the modulus function becomes utterly
trivial. Implement and test the polar version of the class.
Wolves and Caribou
On a certain island there live some wolves and some caribou. In year n there are
obviously wn wolves and cn caribou. What happens the next year depends. . .
• Wolves hunt, and the total number of dinners they get is proportional to
wn kn . The number of baby wolves is automatically proportional to the number of dinners their (potential) parents are able to eat over and above the
amount needed to keep the parents active. The wolf minimum feed intake
and reproductive capability may be modelled as
wn+1 = wn + k1wn (cn − k2)
• In each year the stock of caribou would increase by a factor k3 were it not
for the depredations of the wolves, since each dinner for a wolf is one less
member of the herd. Thus
cn+1 = k3 cn − k1 wn cn
• Baby wolves eat, hunt and reproduce as from year n + 1, and there are no
losses of caribou other than as described above. In particular we do not have
to worry about over-grazing etc.
At the beginning of time the island is stocked with a herd of 100000 caribou,
and a medium-sized pack of ravening wolves. Over a number of years various
things could happen. Either wolves or caribou or both could die out, or the populations could stabilise. For some values of the constants and initial wolf population various of these do indeed occur. For instance if at the start there are twice as
many wolves as caribou the next year there will only be wolves left (one should
adjust the equations given so that negative populations get turned into zero ones),
and the year after that the wolves all expire of hunger.
In fact for many configurations the populations do not stabilise, but they do often get locked into stable cycles that last several years. This improbable situation
has been observed by real naturalists not only in the situation described here but
with regard to disease spread (mumps and children say) and other natural systems.
Write java code to investigate.
110
CHAPTER 4. BASIC USE OF JAVA
Packages and jar files
Make a sub-directory called (say) ex251 and put some Java source files there. Put
package ex251 at the top of the files. Now compile the code, eg saying javac
ex251/*.java. By unless you explicitly set a classpath Java looks for classes
that are in a given package by using the package name as if it described a chain of
sub-directories down from the current directory. So now set up several different
packages and create files that illustrate the use of protected and others that
fail to compile because you have not made allowance for suitable cross-package
visibility. Now look up about jar files and prove to yourself that you can take
a complete Java program (consisting of many classes) and consolidate it into a
single (jar) file that can then let anybody else run it in a simple and convenient
way.
These activities are not essential for any of the example programs that you
have to write for this year’s Java course, but starting to investigate and practise
now will put you in a good position for some of next year’s work, and particularly
the Group Project. I am also aware that this exercise is asking you to read ahead
in these notes. . .
Tickable Exercise 6
The following definition of a paint method uses the rudimentary Complex class
as shown earlier in this section.
The Mandelbrot set can be drawn
by considering the sequence defined
by z0 = 0 and zn+1 = z2n + c where
both z and c are complex numbers.
For most values of c eventually values of zn become large. If one counts
and finds the smallest n such that
|zn | > K for some suitable K then
that n will depend on the value of c
that was used. The well-known pictures arise by using different colours
to display the values of n associated
with values of c = x + iy as x and
y vary. Because drawing this involves a significant calculation for
every single point within the applet’s window it can be painfully
time-consuming. To arrange that the
screen looks more interesting this
4.5. JAVA CLASSES AND PACKAGES
111
code arranges to draw a crude blocky version first and then gradually refine it
into the correct high-resolution image. You may have seen some web browsers do
similar things to give better apparent responsiveness when loading and displaying
pictures from web sites! The code draws a part of the Mandelbrot set centred
around (midX, midY) and with width range, these referring to the values of the
constant c in the iteration. If the value of z has not grown large within LIMIT
steps it is supposed that it never will. The code illustrates use of the Color class.
Colours are sometimes specified in terms of the amounts of Red, Blue and Green
that go to make them up. Printers will tend to think in terms of Cyan, Magenta
and Yellow41 while in yet other circumstances one uses Hue (running through the
colours of the rainbow), Saturation (eg white through pinks up to a full-blooded
rich red) and Brightness (all colours fading to black at zero brightness, just as all
wash out to white (or grey) at zero saturation).
Insert this program in a suitable framework and investigate other areas of the
display by altering the relevant variables. You should be aware that if you increase
the screen size or LIMIT the code can become very time-consuming. Indeed it
might very well be sensible while testing to decrease the finest resolution used to
say 8 rather than 1. And because the paint method computes the whole picture
each time it is called any disturbance of the screen is liable to provoke a complete
re-calculation (at great cost). I find that the appletviewer does not exit until the
end of a call to paint() and so even quitting from it can involve an amazingly
long delay!
public void paint(Graphics g)
{
// I Paint first in crude 16*16 blocks and then
// in finer and finer tiles. This is so that
// SOMETHING appears on the screen rather rapidly.
for (int resolution=16; resolution>=1; resolution/=2)
{
double midX = -0.25, midY = 0.85; // Adjust these
double range = 0.004;
// Adjust this
int screenSize = 400;
// Match .html
int s2 = screenSize/2;
for (int y=0; y<screenSize; y+=resolution)
for (int x=0; x<screenSize; x+=resolution)
{
int n = 0;
int LIMIT = 250;
// Maybe adjust this?
Complex z = new Complex(0.0, 0.0);
Complex c =
new Complex((range*(x-s2))/s2 + midX,
(range*(y-s2))/s2 + midY);
// Important loop follows.
41 Printing inks
favour analysis in terms of subtractive colours rather than additive ones.
CHAPTER 4. BASIC USE OF JAVA
112
while (n++ < LIMIT && z.modulus() < 4.0)
{
z = z.times(z); // z = z * z;
z = z.plus(c);
// z = z + c;
}
// Draw in black if count overflowed
if (n >= LIMIT) g.setColor(Color.black);
// ... otherwise select a colo(u)r based on
// the Hue/Saturation/Brightness colour model.
// This gives me a nice rainbow effect. If
// your display only supports 256 (or fewer)
// colours it will not be so good.
else g.setColor(Color.getHSBColor(
// cycle HUE as n goes from 0 to 64
(float)(n % 64)/64.0f,
// vary saturation from 0.2 to 1.0 as n varies
(float)(0.6+0.4*
Math.cos((double)n/40.0)),
// leave brightness at 1.0
1.0f));
// screen coords point y downwards, so flip to
// agree with norman human conventions.
g.fillRect(x, screenSize-y,
// posn
resolution, resolution); // size
}
}
}
Complete the program based on the above and test it.
Next check Graphics.getClipBounds and Rectangle.contains in the
Java documentation. Adjust the program so that when paint is called it first finds
the clip rectangle associated with the re-paint operation. This is a rectangle on the
screen such that only points within this area need to be re-displayed. Arrange
that the loop on x and y that at present re-computes the colour for every point on
the whole screen just loops round doing nothing for points outside the clipping
rectangle and so only does the expensive operations for points inside it. Try the
new version, and in particular move other windows to obscure small parts of it and
then move them away so you can see the effect of the partial re-draw operations.
Note that the above program will display best if your screen is set up to support
lots of colours. On a display with either 16-bit colour (65536 colours) or truecolour (24 or 32 bit) and at high resolution the effect is fairly stunning. If only
256 colours are supported the shapes will remain nice and wiggly but the delicate
shading will be lost. While preparing these notes I have adjusted the program
4.5. JAVA CLASSES AND PACKAGES
113
to display a 1200 by 1200 image at best-possible resolution in 16-bit colour, and
although it takes utterly ages for the screen to refresh I think it is almost worth it!
The program that I give has a bug that you can see if you watch carefully when
it re-paints at the various different resolutions. It relates to the fact that in Java the
x-co-ordinate increases from left to right (as expected) but the y-co-ordinate is
zero at the top of the screen and largest at the bottom. Identify and correct the
behaviour that I count as a defect.
Optional: Add a BufferedImage to make the re-painting of the screen cleaner. I
might like to be able to reset the view to some standard one at the click of a mouse,
and to be able to drag with the mouse to select a sub-part of the current picture for
zooming in on. Those who are feeling keen can investigate these possibilities.
There is also quite some incentive to find ways of speeding up drawing of the
images here!
(End of tickable exercise)
Fractions
Create a class similar to the Complex one but that implements rational numbers,
is fractions. You will probably want to make the internal representation a pair of
long values rather than just int, and keep everything reduced to lowest terms by
cancelling out highest common factors.
Series for tan(x)
It may be well known that
1
2
17 7
62 9
tan(x) = x + x3 + x5 +
x +
x +...
3
15
315
2835
but fewer people are happy about being able to predict what the next few coefficients in the expansion are. However if we have a computer program able to
compute with rational numbers it is in fact easy to generate as many more coefficients as are desired. The coefficients satisfy a recurrence formula
t0 = 0
t1 = 1
1 n−1
tn =
∑ titn−i−1
n i=0
Use this to confirm the series as I have tabulated it and display the next few
terms. The result here may be derived from the fact that the derivative of tan(x)
is 1 + tan2 (x), and is also related to (but rather harder than!) the discussion of
“generating functions” in the probability course.
CHAPTER 4. BASIC USE OF JAVA
114
Complex elementary functions
Perhaps you already did this when making your complete version of the complex
numbers class. . .
It might be useful to be able to construct complex numbers either by specifying
real and imaginary parts or by giving argument and modulus42. Failing any better
scheme you could distinguish between the two constructors by adding declarations
public static final int CARTESIAN = 0;
public static final int POLAR
= 1;
in the Complex class and then having the constructor take an extra argument that
specifes which option is being used. It would then also make sense to provide
data access methods that make it equally easy to access the number in polar or
cartesian interpretation.
If that is done it becomes reasonably easy to support complex versions of
several of the elementary functions. Observe the identities:
p
√ iθ/2
reiθ =
re
iθ
log(re ) = log(r) + iθ
exp(x + iy) = exp(x)eiy
sin(z) = (exp(iz) − exp(−iz))/2i
cos(z) = (exp(iz) + exp(−iz))/2
pq = exp(q log(p))
The expressions for sin and cos can be inverted to find ways of writing the inverse
trigonometric functions as messy complex logarithms. And it may be seen that the
neatest way of using these formulae to implement complex-values versions of the
elementary functions really does benefit from being able to slip very comfortably
between the cartesian and polar views of the values. Implement it all.
I should observe carefully that the code you have just written is liable to be a
very long way from the last word in elementary function libraries, for the following reasons, which are given in descending order of importance:
1. Several of the complex-valued elementary functions have branch-cuts. For
instance the square root function has a principal value which is discontinuous as you cross the negative real axis, and the various inverse trig functions
will also have cuts. Your code can not automatically be assumed to implement these cuts in the way that will be considered proper by experts in the
field. Probably the most readily accessible description of which cuts are
desirable is in Common Lisp, the Language by Guy Steele[21];
42 Ie
by giving the polar version.
4.6. INHERITANCE
115
2. Your implementation will probably suffer from arithmetic overflow (and
hence give back answers that are infinities or NaNs) substantially before
the desired result would overflow. For instance the identity given for cos
computes an intermediate result that is twice as big as the final answer, and
hence can suffer in this way thereby returning incorrect answers;
3. In many cases the naive use of the formula given can lead to serious loss
of numerical accuracy when values of similar magnitude are subtracted one
from the other. For instance this problem would arise in the calculation of
sin(z) for z near zero;
4. Direct use of these formulae will not even give an efficient set of recipes for
the desired functions!
however the numerical analysis to address these problems is certainly beyond the
scope of this course.
Binary Trees
Start from the BinaryTreeOfIntegers class sketched above and extend it so
that as well as defining variables in the class it provides a set of methods to work
with them. The methods you introduce should arrange that any binary tree built
is always structured so that all integers stored in the left sub-tree that hangs off a
node are smaller than the integer in the node itself, while all integers in the right
tree are greater (or equal). You should provide a constructor that creates such
a tree out of all the integers in an integer array, and another function that first
counts the size of a tree, then allocates an array that big and finally copies all the
integers back into the array so that they end up in ascending order. I would fairly
strongly suggest that you design and implement the key parts of this in ML before
you move on to the Java version. Your code is an implementation of tree-sort:
you should compare it with quicksort for clarity, amount of code that has to be
written, robustness (ie are there any truly bad sorts of input it can be given) and
performance.
4.6 Inheritance
There is one more major feature of the Java class mechanism. It provides yet
further refined control over name visibility and it can often be a huge help when
organising the structure of large projects. It is called inheritance and the idea of it
is to allow the user to define new classes as variants on existing ones. When this
happens the new class starts off with all the components and methods of the one
116
CHAPTER 4. BASIC USE OF JAVA
upon which it is based, and it counts as having defined a sub-type. It can however
define extra variables and/or methods and implement more specialised versions
of some of the methods already present in its parent class. This is what has been
happening every time we have used the word extends, and so for instance every
applet we have written has defined a new class extending the library class Applet.
This library class implements all the major functionality for getting a window to
appear, and to get the visual effects we wanted all that was needed was to provide
our sub-class with its own version of a paint method.
There seem to be three interlocking reasons why inheritance is important when
large programs are to be written:
1. Class libraries can be provided in forms that implement all the generic behaviour of really quite complicated programs, but by making a new program
that inherits from such a class and that overrides some of its methods lots
of flexibility is left for the programmer to create a system that does exactly
what they want. Prior to languages that supported inheritance there was a
severe conflict between having libraries that contained large enough components to give large time-savings and those that were adaptable enough to
be realistically useful;
2. Class inheritance serves a linguistic purpose in Java. If you start from a single base class it is possible to derive several other classes from it. All these
count as specialisations of the original one, and a variable capable of holding a member of the base class can therefore automatically refer to instances
of any of the derived ones. This is how Java can support data-structures that
can have several variants. Furthermore the name-visibility rules in Java can
use the way in which inheritance groups classes into families to further refine access to class members.
3. It often becomes possible to implement a set of basic classes first, and test
them, and then leave those alone (and hence stable) while deriving new
classes that add extra functionality. This both provides a respectable strategy for organising system development, and means that there is a significant
chance that the basic classes that are developed will be useful in the next
project;
I will try to illustrate these three points in turn.
4.6.1 Inheritance and the standard libraries
The richest and most valuable place where this happens in the libraries relate to
applications that pop up windows. Examples given before show user code being
4.6. INHERITANCE
117
derived from a class called Applet. One of the things that has been seen about
Applet and hence any class derived from it is that the method paint has a special
status, in that it is invoked whenever the screen needs to be refreshed. The fact that
by deriving a new class you get an opportunity to write your own paint method
and that in your new class your own definition takes the place of a standard one
(which probably does nothing much!) is obviously critical. If you could not alter
the re-painting behaviour of an applet the whole structure would lose its point. If
you look at the documentation for the Applet class you will find that it is listed
as having around a couple of dozen associated methods. Each of these will define
a default behaviour for an Applet and each can be replaced43 in a derived class if
some special behaviour is needed. However these two-dozen methods are very far
from being the whole story. For instance paint is not listed among them. This is
because Applet is descended from java.awt.Panel which in turn is derived
from java.awt.Container which itself inherits from java.awt.Component
and java.lang.Object. Each of these super-classes define (often many) methods of their own. The lower-down ones sometimes replace a few of the higher
level methods with more specialised versions, but they also tend to provide lots
of new methods of their own. Thus in this case the paint method is defined as
an aspect of a Container, and is only part of Applet via inheritance. The end
effect is that something that is as easy to get started with as an Applet in fact
comes complete with perhaps hundreds of bits of pre-defined behaviour almost
any of which can be adjusted by the simple expedient of overriding some method.
Sometimes of course this arrangement whereby library facilities are structured
into hierarchies of classes means that the very simplest thing one might want to do
involves explicit construction of objects from various classes in a way that looks
less smooth. To print simple text as the output from a simple Java stand-alone
application one can invoke System.out.println. The long name is because
System is a class (its full name is java.lang.System), and out is then a variable in that class. The field out has as its type PrintStream and the class
PrintStream provides a method called println. It is possible to reference the
variable out just by giving its class (without having to have a variable whose type
is that class) because it was defined as being static. The recipe as typed in by
the programmer is not too bulky but the full explanation of why it works is a bit
clumsy. “Simple” input is if anything worse. There is a static variable System.in
which is of type InputStream, and for an application to accept input from the
keyboard one needs to use it. However the class InputStream only provides the
most basic reading functions, and various derived classes are needed if flexible,
efficient and convenient reading is to occur. A suggested protocol for a single
43 The
fuller story is that any member of a class that has been marked as final can not be
redefined in a derived class. The use of final thus provides the designer of a class with a way
to guarantee some aspects of class behaviour even in derived classes.
CHAPTER 4. BASIC USE OF JAVA
118
integer from the standard input ends up something like
BufferedReader in =
new BufferedReader(
new InputStreamReader(System.in),
1);
int n;
try
{
n = Integer.parseInt(in.readLine());
}
catch (IOException e)
{ n = -1; }
catch (NumberFormatException e)
{ n = 0; }
System.out.println("I got: " + n + "....");
This creates an InputStreamReader out of System.in, and then builds from
that a BufferedReader where here I have indicated that a buffer size of 1 should
be used. For reading directly from the keyboard a ridiculously small buffer size
means that the program gets characters as soon as they are available. If the “, 1”
was omitted the BufferedReader would use some default buffer size and you
would have to have keyed in that many characters before anything ever happened!
The BufferedReader class then provides a readLine method, and the string
that it returns can be interpreted as an integer by the static method parseInt in
the Integer class. Both readLine and parseInt may raise exceptions if anything goes wrong, and so a proper program should be prepared to handle these.
The above tends to look very heavy-handed because “real” programs will generally want to decode much more complicated input than just the single number
shown above, and will really need to put in the catch clauses so that they can
respond cleanly to erroneous input. Even the buffering control is really quite important — direct keyboard input may need to be unbuffered so that interaction
works well while input of large amounts of input from a file may be much faster
if buffering is used.
Java in fact provides another rather larger class than BufferedReader which
may be useful in many applications that want to accept free-format input. This is
the class java.io.StreamTokenizer44 which can help you read in a mixture
of numbers and words. Here is a demonstration:
import java.io.*;
44 Actually
I think that StreamTokenizer is very useful while you are getting started, but
although it can be customised quite substantially it is not flexible enough for most really serious
uses. In the Compiler Construction course in Part IB you may learn about a package called JLex
that is harder to set up but which provides enormously more power and flexibility.
4.6. INHERITANCE
...
StreamTokenizer in =
new StreamTokenizer(
new BufferedReader(
new InputStreamReader(System.in),
1));
in.eolIsSignificant(true); // see newlines
in.ordinaryChar(’/’);
// ’/’ is not special
in.slashSlashComments(true); // ’//’ for comment
try
{
int type;
// The next line loops reading tokens until end of file.
while ((type = in.nextToken()) !=
StreamTokenizer.TT_EOF)
{
switch (type)
{
// There are a number of predefined "token types" in
// StreamTokenizer, so I process each of them.
case StreamTokenizer.TT_WORD:
System.out.println("word " + in.sval);
// If the user says "quit" then do so. NB "break" only
// exits the switch statement here.
if (in.sval.equalsIgnoreCase("quit"))
break;
continue;
// in.sval and in.nval get set when string or numeric
// tokens are parsed and contain the value.
case StreamTokenizer.TT_NUMBER:
System.out.println("number " + in.nval);
continue;
// the method lineno() tells us which line we are on.
case StreamTokenizer.TT_EOL:
System.out.println("start of line " +
in.lineno());
continue;
// quotes and doublequotes contain strings.
case ’\’’: // drop through
case ’\"’:
System.out.println("string " + in.sval);
continue;
// Other characters end up here. Eg +, - etc.
default:
119
CHAPTER 4. BASIC USE OF JAVA
120
System.out.println("sym " + (char)type);
continue;
}
break;
// here if "quit" typed in
}
}
catch (IOException e)
{
System.out.println("IO exception");
}
The level of complexity here seems much more reasonable! The initial code that
sets up a StreamTokenizer is not very different from that which set up the
simpler buffered stream before, and is clearly a small overhead to pay to be able
to have Java split your input up into words and numbers. The StreamTokenizer
provides methods that allow you to customise its behaviour so that it can recognise
one of several possible styles of comments and accept various string delimiters.
The calls
in.eolIsSignificant(true);
// see newlines
in.ordinaryChar(’/’);
// ’/’ is not special
in.slashSlashComments(true); // ’//’ for comment
illustrate a little of this. The first call tells the tokenizer that newlines should
be returned to the caller. By default they are counted as whitespace and so not
passed back. The second call makes a single / into an ordinary character, where
by default it introduces a comment if followed by a second / or a *. The final line
enables recognition of comments that are started by //. As always you need to
browse the full documentation to discover what all the other options are!
Two lessons emerge. The first is that the bigger and more powerful classes
in the Java libraries may really save you time if you find out how to use them,
while direct use of very low level facilities may end up feeling pretty clumsy.
The other is that these high level facilities are often very flexible, but if you need
some feature that they do not support you may have to drop down a level. For
instance StreamTokenizer does not know how to handle numbers expressed in
hexadecimal or octal, and it always reads numbers in type double which is not
good enough if what you needed was a long value.
4.6.2 Name-spaces and classes
When you derive one class from another it is sometimes desirable if the methods
and fields of the base class are visible in the derived one, but in other cases it may
not be. This aspect of name visibility needs to be considered in conjunction with
the consequences of classes falling into different packages. Java confronts all this
by defining four levels of name visibility within classes:
4.6. INHERITANCE
Figure 4.6: Classes and inheritance are a sort of magic.
121
CHAPTER 4. BASIC USE OF JAVA
122
private: is the most restrictive one. A method or variable that has been declared as private can be referenced from within the class in which it is
defined, but not from anywhere else. In particular code that is in another
class can not see it regardless of whether the other class is in the same package as or was derived from the original one;
package: relaxes things so that code in any class that is in the same package can
reference a value. This is the default arrangement, and is indicated by not
using any of the other visibility qualifiers. Note that the keyword package
is used at the head of a file to specify which package that class will reside
in, and it is not valid in method or variable declarations;
protected: When a name is declared as protected it becomes visible in
derived classes even if they are in other packages. Because during this first
course you will probably not be creating new packages yourself this case
will mostly be relevant where a library class has some protected members
and you derive a few class from it. Your class will probably be in the default
package but despite that you will be able to access the members involved;
public: is the final case, and it makes names generally available regardless of
packages and inheritance.
It seems tidy to document the other possible qualifiers for declarations here,
even though they are not concerned with name visibility. Indeed their consequences are rather mixed, and since this is a first Java course it is not essential to
be fully comfortable with them all.
final: When a variable is declared final nobody will be allowed to assign
a new value to it. When a method is final then it can not be overridden
in any derived class. In both cases the effect is to make the definition in its
visible form the one that can be relied upon everywhere else;
static: The default situation for items defined within classes is that the items
only come into existence when an object of the class-type is created. This
makes obvious sense for data fields. For instance after the declaration
class IntList
{
public int head;
public IntList tail;
}
it is clear that the only context in which the head and tail fields can be used
is in association with an object of type IntList as in
4.6. INHERITANCE
123
int sum(IntList x)
{
int r = 0;
while (x != null)
{
r += x.head;
x = x.tail;
}
}
For consistency the same access rule is then applied to member functions
(ie methods) in a class. If however an item in a class has been declared
static it is as if a single globally allocated instance of the class gets created automatically, and the field can then be referred to relative to just the
class name. For instance (a nonsense code fragment!)
class MyConsts
{
static final double ZETA2 =
1.6449340668482264365;
static final double CATALAN =
0.91596559417721901505;
static int square(int x)
{
return x*x; }
}
...
double a = MyConsts.CATALAN Myconsts.ZETA2 +
(double)MyConsts.square(1729);
...
abstract: Sometimes it is useful to define a base class not because it is useful
as such, but because the various other classes that get derived from it might
be. Consider the ML declaration
datatype option = A of int | B of double;
One way of producing a Java equivalent would be to start by defining a
rather vacuous class called Option and then deriving from it two new
classes one to correspond to each of the two cases in the ML version:
abstract class Option
{}
class OptionA extends Option
{
int a;
}
124
CHAPTER 4. BASIC USE OF JAVA
class OptionB extends Option
{
double d;
}
The base class here only exists to be extended, and it would be silly to create
an object that was of that type45. The qualifier abstract prevents anybody
from creating objects of the base class. It marks things that must be inherited
from before meaningful use can be made of them. In cases such as this it is
often useful to discriminate as to which derived class a particular instance
belongs to. The instanceof operator can be used to do this. Again my
illustrative code is artificial:
Option x = new OptionA(); // or maybe OptionB?
...
if (x instanceof OptionB) ...
else ...
It is very often neater and easier to define different overridings of a common (abstract) method in the two derived classes so that the correct behaviour is achieved for each. If that is done46 the if statement and use of
instanceof could be replaced by a simple call to the method concerned. It
is of course not essential to make a base class in such examples abstract,
but doing so prevents any possible embarrassment if some code created an
instance of it in its raw and useless form, so it is generally considered to be
good style.
native: If a method is defined as native then Java somehow expects there to
be an implementation of it that was coded in some language other than Java.
This can be used by system builders to interface Java code down to lower
level and perhaps machine-specific system calls, but will not be discussed
further in these notes.
synchronized: related to Java code where several threads of computation may
be active at once. Although the very basic aspects of this will be covered
in this course a proper treatment needs to wait until you have had a Part IB
course on concurrent systems.
interface: The keyword interface is not a modifier for use in class definitions but a keyword whose use is very much like that of class. An interface
45 Of
course objects of type OptionA and OptionB are also of type Option, so what I
mean is it would be silly to go new Option().
46 A similar stylistic issue arises in ML where user of pattern-matching in function definitions
can often reduce the number of explicit if statements that have to be written.
4.6. INHERITANCE
125
can be declared much as an abstract class is. Classes can be defined to extend other classes, but a restriction that Java applies is that a new class can
only be an extension of a single parent class. Interfaces provide an approximation to being able to extend several parent classes — a new class can
specify that it implements one or more interfaces. When a class indicates
that it will implement an interface it has to contain (concrete) definitions of
all the (abstract) methods that the interface specifies.
At (very) long last we have covered all the magic that arose in the initial
Hello.java program and can see what each keyword present there was indi-
cating.
4.6.3 Program development with classes
In Java, as in other Object Oriented languages, the whole shape of a large program needs to be designed in terms of terms of the packages and classes that will
be built. It is worth putting particularly careful thought into the way in which
hierarchies of classes will be derived from one another via inheritance.
There are two application areas that were pioneers in illustrating the benefits
and strengths of object oriented programming (which is what this is). It can thus
be worthwhile considering examples of these as some of the earliest ones you
work with when getting used to the idiom. The first application area was that
of simulation47, while the other was graphics and especially the display of geometric figures in windows. The following example, which is taken from Java in
a Nutshell and shows how use of several classes rather than just one may allow
the programmer to keep distinct aspects of their task separate. But doing this the
size of unit that has to be debugged is reduced, and the possibility of re-using
parts of the code later on in another project is increased. The example supposes
that a graphical design and modelling package is being written. Within it it will
keep data-structures that represent circles, squares and other shapes. For much of
its time it will work on these busily computing their areas, their circumferences,
whether they intersect and similar properties. It may also adjust their sizes and
positions. As well as performing all these calculations the complete package will
also have a user interface that can draw the objects. There will be options to control the colour of each individual circle (and so on) as well as to determine whether
the items are drawn just as outline figures or as filled-in shapes.
Without use of inheritance and thus without serious use of the Java class mechanism the code would probably have to consist of a single class, say called Shape,
which would contain a master variable indicating what sort of shape was involved,
47 Indeed
the way that object-oriented C++ developed from the simpler language C was initially
specifically for use in this area.
126
CHAPTER 4. BASIC USE OF JAVA
Figure 4.7: See also the “Software Engineering” courses.
4.6. INHERITANCE
127
then other variables that could be used to specify the exact parameters of that
shape (eg its radius if it was a circle). The method functions such as area would
need to dispatch on the type of the figure and do different calculations in each
case. Further code would arrange to be able to draw pictures to represent the data.
All the geometric and graphical parts of the code would be in the same class and
thus the same source file — something which would not cause trouble in tiny cases
but would become clumsy for a fully elaborated version.
With inheritance it would be natural to start with a basic class (again I will call
it Shape) which will probably be abstract. Its purpose is to allow the program
to declare variable of type Shape and then store circles, squares, stars and all
other possible sorts of shape in that single sort of variable. The methods declared
for Shape can be given as just declarations, rather than as full definitions:
public abstract class Shape
{
public abstract double area();
public abstract double circumference();
}
which makes these methods available in any object of class Shape but expects
that concrete variants on the class will provide the real implementations.
For each sub-class of Shape a new class could then be derived:
class Circle extends Shape
{
protected double radius;
public Circle() { radius = 1.0; }
public area() { return Math.PI*radius*radius; }
public circumference()
{
return 2.0*Maths.PI*radius; }
public double getRadius() { return radius; }
}
Note that this can introduce new public members that are not relevant for general
Shape quantities, but which do make sense when you know you have a Circle.
Next an interface would be set up, defining the methods relevant for draw48
ing things on the screen:
48 Java
is an American language, and so the character of being Red, White or Blue is Color
rather than Colour. Given that the library uses this spelling it seems best to swallow nationalistic
pride and adopt it elsewhere in the code. . .
128
CHAPTER 4. BASIC USE OF JAVA
public interface Drawable
{
public void setColor(Color c);
public void draw(Graphics g);
// etc etc.
}
Now it is reasonable to derive a new class for a version of each sort of shape but
in a form that supports the drawing operations:
class DrawableCircle extends Circle
implements Drawable
{
Color c;
public void setColor(Color c)
{
this.c = c; }
public void draw(Graphics g)
{
... // whatever, maybe
g.drawOval(...);
}
// etc etc.
}
It is now possible to use the drawing methods as well as the data manipulation
methods in one of these ultimate data-structures.
Often when producing a derived class and overriding a method the newly extended method needs to use the corresponding operation from its parent class. For
instance if a class defines a method that is used to initialise its variables then a
derived class may add extra variables that need initial values too, but it would be
clumsy to insist that it also had to repeat all the code to setup the variables in the
base class. And indeed if some of those were private or protected it might
not be able to. The solution is hidden in the keyword super. This is a bit like
this in that it always refers to the current object, but it views it as a member of
the immediate parent class. Thus code like
class SubClass extends MyClass
{
private int variable;
public void init()
{
super.init(); // init as a MyClass
variable = -1; // finish off as SubClass
}
}
and the word super is only of relevance when extending a class and overriding
methods. In the case of some library classes and methods the documentation will
4.7. GENERICS
129
explain to you that you must use it, see for instance the method paint in the class
Container.
4.7 Generics
The material here is now for Java 1.5 and I expect my coverage of it to grow over
the next year or so. This year I will do hardly more than just mention it and let
Part IbB coverage consider filling in the gaps. This seems especially reasonable
since textbooks that catch up with this are still somewhat rare.
In ML you got used to having types that were polymorphic. For instance a
sort function that took a predicate and a list might have had type
(α ∗ α → bool) ∗ αlist → αlist
to indicate that the elements of the input and output lists had the same type and
the ordering predicate was compatible with that. A particular feature of ML to
recall is the availability of parameterised types such as αlist. In Java instead of
saying “type” we will say “class”, and instead of saying “polymorphic” we say
“generic”. A generic class is established by putting type variables within angle
brackets. You can then use the type variable within the class as if it were a regular
type name: small
class MyClass<E>
{
E myMethod(E arg1, int arg2)
{ MyClass<String> newvar = ...
...
}
}
With your ML experience of polymorphism you can now probable see at once
how to use this capability to write implementations of various generic data structures (trees, lists and the like) and provide useful functions that traverse, search or
sort them. In fact that Java libraries have done a great deal of that in their so-called
Collection classes.
In ML polymorphism is all-or-nothing. If you have a type-variable α it can
stand for absolutely any ML type. To improve security you may sometimes like
to have a way of expressing more limited flexibility (eg generic over all sorts of
numbers, but not over non-numeric data). Java provides a capability using a type
wildcard written as question mark, and can limit the range of the wildcard using
notation like
CHAPTER 4. BASIC USE OF JAVA
130
public void sum(List<? extends Number> arg)
{ for (n:arg)
{ ... }
}
Here the sum method takes an argument that is some sort of List49 but it
insists that the polymorphism that List provides has been used in a way that
means you know that all the objects in the list are some subclass on Number.
You will use generics every time you use the Java Collection Classes. You can
use it in your own code too. There is a fair amount more that I could say about
exactly how it interacts with the type-hierarchy that class inheritance provides and
when a generic class is a sub-class of another, but I believe that the details there
do not belong in a first Java course!
4.7.1 Exercises
Objects everywhere
The Java libraries make extensive use of classes in hierarchies (and also a more
modest number of interfaces). The arrangement in the basic set of classes is that
everything is ultimately descended from a base class called Object. The most
immediate consequence is that an object of any class from the basic libraries may
be stored in a variable of type Object. It is exactly as if whenever you define
a new class and do not give an explicit extends clause as part of its definition
Java just sticks in “extends Object” for you. Of course when you extend some
other class it in turn will somehow have Object as an ancestor-class so this way
as previously stated every instance of any class is an Object.
A few basic methods are defined for Object, of which perhaps the most interesting at present is getClass which returns an thing from the class Class. If x is
any Object then x.getClass().getName() is a string that is the name of the
class of x! The general parts of the Java libraries that allow you to investigate the
classes that Objects belong to and then retrieve lists of the variables and methods
that they provide are referred to as Reflection: as it were a Java program can look
at itself as if in a mirror.
Check the documentation and write Java code that accepts an Object and
prints out as detailed and as readable description of it as you reasonably can.
Note that Object underpins the polymorphism of Java generics, but now that
generics are available programmers will use Object directly much less than they
used to.
49 A
Collection class that does just what you expect!
4.7. GENERICS
131
Primitive is second class?
The ability to treat things as “Objects” does not (directly) extend to the Java primitive types. To work around that the libraries contain classes with names that
are rather like those of the primitive types except that they are capitalised. Ie
Boolean, Character, Byte, Short, Integer, Long, Float and Double. As
of the most recent revision of Java you will find that the compiler arranges to convert between int and Integer (and the other primitive types and their associated
wrapper classes) when it believes that that will help you. The conversion naturally
involves some run-time cost so it is perhaps advisable to be aware when it happens. The sort of circumstance where it is especially convenient that this happens
is when you want to store a primitive object (eg an integer, floating point number
or character) in a Hashmap or a Vector (or indeed any of the collection classes).
It was then natural for the Java designers to set methods associated with these
to implement a wide range of basic conversions and tests on the values, as in
Integer.doubleValue and Double.isNaN (and many more).
The numeric types the classes Integer etc do not inherit directly from Object
but via a class called Number Eg
Number a, b;
a = 2;
// new Integer(2);
b = 11.7; // new Double(11.7);
can be written as shown, but behaves as if the constructors in the comments have
been used. If the Number objects are used in a context where primitive numbers are needed (eg you try to perform arithmetic using them) the values will be
unpackaged for you.
Write a class that defines lists of Numbers, with suitable set of facilities for
constructing such lists and a method sum that can add up the values in a list returning the result as a double. You may need to use “x instanceof Integer” to
sort out which flavour of number is present in some particular node.
Some text output using Objects
Since Object is an almost universal type it can be used to pass arbitrary data
to a function. This is in fact what happens with printf, but one extra thing
happens there. If a method is declared with three dots after the type at the end of
its argument list, as in
PrintStream printf(String format, Object... args)
{ // whatever definition you need
}
132
CHAPTER 4. BASIC USE OF JAVA
indicates that the final argument to printf will actually be passed as an array of
Object values. But the calls to it will just appear to permit a variable number
of arguments, and each argument will be converted to (if a primitive type) or
interpreted as (if a class type) Object.
While this scheme can be used in your own code to support variable numbers
of arguments, and it can also be used with more restrictive types than Object it
will almost always count as poor style since it can easily reduce type-safety and
cause confusion if you mix it with method overloading. But where it is useful it
really helps make code concise. Without it instead of writing
int i=1, j=2;
System.out.printf("%d, %d", i, j);
you would need to wrap i and j up in the type Integer explicitly, and create an
array to pass the multiple arguments explicitly.
int i=1, j=2;
myOwnMethod("%d, %d",
new Object [] {new Integer(i), new Integer(j)});
But note very well that within the code that implements things such as printf
everything has to work understanding that the concise calls are in fact mapped by
the java compiler onto the clumsy looking code that packaged up primitive types
and makes an array.
Now seems a good time to provide a summary of more of the formatting options available with printf, and also to note that the method String.format
does exactly the same job of layout but returns a formatted string rather than doing
any direct printing. We have already seen "%d" for laying out integers, and know
that "%n" generates a newline.
Within a format string the character "%" introduces a format specifier. After
the percent sign a number of optional elements can appear:
• An argument index followed by a dollar sign ($). Without one of these the
values to be converted are taken one at a time from the arguments provided.
An index such as (2$) tells the formatter to use the second argument now,
even if that is out of order. Often you may want to display the same data
several times, eg in different formats. In that case (<) is very useful: it tells
the system to re-use the argument most recently dealt with;
• Some flag characters. Just what is valid here will depend on just what sort
of layout is being performed, but various punctuation marks as flags can, for
instance, force left-justification of text within a field (-), ensure that numbers are always displayed with an explicit sign (+), include leading zeros
(0) or be more fussy about the actual types of arguments (#). You need to
check fine details in the documentation when you use flags!
4.7. GENERICS
133
• a field-width, written as an unsigned non-zero integer. You should expect
that if this is specified that the output from the conversion will have exactly
that number of characters;
• A dot followed by a integer precision. Eg (.4). For some conversions this
sets an upper limit on the number of characters to be generated. For floating
point conversions it controls either significant figures of the number of digits
after the decimal point.
• (and finally!) a character (or in some cases a pair of characters) that indicated just what sort of conversion is to be performed. Perhaps the more
important cases are the letters s, each of which is discussed briefly below!
The full set of format letters and options can be found in the online documentation, but key cases are
s, S: This takes any value at all and tried to convert it] to a string. If the argument implements the Formattable interface then its formatTo method is
used to do the conversion, otherwise its smalltoString method is used.
When you define a class of your own you may often wish to override or
define one or both of these methods so that you can easily print instances
of your class. Many of the Java library classes implement these methods in
ways that at least try to be helpful. If you write a capital S the material that
is displayed is forced into upper case. Similar effects apple for other use of
upper case format letters;
d: This is the case most often seen in these notes so far, and prints an integer. But
you can also display BigInteger values with this (and the x) format;
x, X: Integers can be displayed in hexadecimal rather than normal decimal notation this way;
c: character
e, f, g, E, F, G: Floating point and their display involve lots of complication! The “e” formats always use scientific notation with an explicit exponent. The “f” formats use the specified precision as the number of digits
to display after the decimal point (eg it is a good thing to use for printing
pounds and pence with "%.2f"), while “g” tries to select between those
two formats to select one that will be natural and will take up as little space
as possible.
%: If you want to print a percent sign you will need to write two in a row!
CHAPTER 4. BASIC USE OF JAVA
134
n: Unix and Windows have different ideas about what constitutes a “newline”.
The format code %n makes allowance for that for you.
tx: Java provides an amazingly rich range of ways of formatting times and dates.
YOu can use these formats when printing objects of type Long, Calendar
or Date. I think there is too much to list here, but a very few of the options
available are
tY year displayed as 4 digits, eg “2005”;
tA full name of day of the week, eg “Monday”;
TA as above, but upper case: “MONDAY”;
ta short name of day: “Mon”;
tM minute within the hour, as 2 digits;
tm number of the month as 2 digits, counting January as number 1;
tT time formatted for the 24-hour clock as "%tH:%tM:%tS";
tD date formatted as "%tm/%td/%ty".
A bigger exercise
There are twelve shapes that can be made by joining five squares together by their
edges to get a connected unit. It is possible to pack these shapes (the pentominoes)
into a six by ten50 rectangle in a number of ways. Here is one such packing, which
will also serve to show you the shapes of all the pieces:
50 also
into a five by twelve or three by twenty.
4.7. GENERICS
135
The object of this exercise is to find other solutions to the puzzle.
The suggested strategy is to represent the 10 by 6 board using 60 of the 64 bits
in a long. You can them treat these as if they are arranged as a rectangular array,
and then a single long value can represent a possible position of a piece. In this
representation the twelve pieces can be described by the array:
final long [] rawPieces =
{
0x000001f, 0x0100407, 0x000040f, 0x0300403,
0x0401c01, 0x2008007, 0x0201c02, 0x0000c07,
0x0301808, 0x0000c0e, 0x000100f, 0x0301802
};
where the values look pretty ugly but are at least all in quite a small table. A
bulkier but perhaps cleaner way to set up the initial table of shapes would be to
use a function such as:
long piece(String line1, String line2, String line3)
{
return (row(line1) << 2*boardWidth) |
(row(line2) << boardWidth) |
row(line3);
}
long row(String line)
{
long r;
for (int i=0; i<line.length(); i++)
{
r <<= 1;
if (line.charAt(i) == ’X’) r |= 1;
}
}
...
piece("X ",
"XXX",
" X ");
The init method for the applet should start by setting up a table first of all
the twelve pieces normalised so that they are down in one corner of the board,
and then a larger table showing each piece in every location on the board that it
could possibly be. Doing this will involve writing code that reflects and rotates
pieces — not especially nice when using bits packed into a long — and which
avoids setting up entries that are redundant because of symmetry. The code that is
involved in getting this far is quite messy enough to keep you busy for a while.
CHAPTER 4. BASIC USE OF JAVA
136
The overall structure of the code that searches for solutions might then be
//
//
//
//
//
//
//
search(i) looks for ways of placing piece
i on the board. The array entry maps[i][j]
is a bitmap showing the j-th place that
piece i could bit, and the variable "board"
shows which parts of the board have already
been filled. There are 12 pieces, known as
0 to 11.
void search(int i, long board)
{
if (i == 12)
{
// Here a solution has been found
... record it somehow ...
return;
}
for (int trial=0; trial<maps[i].length; trial++)
{
if ((maps[i][trial] & board) == 0)
{
// no overlap with existing pieces
// so put this in and next try to
// fit in piece i+1.
search(i+1, board | maps[i][trial]);
}
}
}
The first challenge would be just to count the solutions, and so the place in
the above which is incomplete could be replaced by a single statement that incremented a variable. But since it is easy to use fillRect to draw filled-in
rectangles in Java it would seen natural to try to draw some of the solutions and
that would mean doing something distinctly harder.
The search function I have sketched tries the twelve pieces one after each
other, and at each stage considers each piece at ever position on the board where it
would still fit. A different search strategy would be to scan the board at each stage
and find the first vacant square. The program would then identify and try every
piece that could be used to fill in that square. I believe that this second search
strategy is rather closer to the one most people would use than my original one
was.
A curve to plot
Use Java to plot a picture of the following curve as t varies from 0 to 2π:
x = cos(t)(1 + cos(40t))
y = sin(t) + cos(t) sin(40t)
4.7. GENERICS
137
Find a copy of A Book of Curves, E. H. Lockwood, Cambridge 1963, and in a
similar style re-create variants on as many of the pictures as you can.
Reading hexadecimal numbers
We have seen various ways of decoding numeric input, eg Integer.parseInt
and the whole set of joys associated with StreamTokenizer. You can note from
the full documentation that there is a two-argument version of parseInt that
allows you to specify what radix the input string was supposed to be in. You
may also like to check details of the class smallScanner which has a method
nextInt that can also accept an argument indicating what radix to read in.
Now imagine that these facilities did not exist, or that for some strange reason
you could not use them. Implement your own functions that can be given strings
as arguments and which will make it possible to convert the strings into int and
long values, allowing for the possibility of octal or hexadecimal specifications.
Displaying floating point numbers
In versions of Java prior to 1.5/5.0 the functionality of printf was not available.
This exercise is to re-create some of it thereby getting a chance to feel what work
is involved in making worthwhile extensions to the existing libraries.
The method Double.toString allows you to generate a printable representation of a floating point number. However compared to the floating point layout
flexibility available in many other languages it seems pathetically simple-minded.
A typical programming language will provide for three ways of printing floating
point values:
F format: here numbers are written as illustrated in the following examples
-1.000
1234567890000000000.0
0.000000005656
and even if the values are very large or very small their magnitude is indicated by having suitable numbers of leading or trailing zeros. It is typically
possible to specify how many digits will be printed following the decimal
point, and to indicate the width that the whole number will be padded to
with either leading or trailing blanks.
E format: For very large or small numbers it may be convenient to use scientific
notation. So with E format an explicit exponent will always be displayed:
CHAPTER 4. BASIC USE OF JAVA
138
-1.0e000
1.234568E018
5.656000e-009
Observe that there is always exactly one digit before the decimal point
(sometimes a scaling option is provided to allow the user to specify a different number of digits before the point), and the exponent is always present
and probably always displayed in a way where the largest possible exponent
value could be fitted in. A “precision” specifier can indicate how many significant digits are to be shown, and the number will be padded with zeros
or rounded to meet that requirement. Numbers close to 1.0 tend to look a
bit ugly this way! Again it is useful to be able to place the number in a
fixed-width field, either right or left-justified.
G format: Large numbers are best shown in E format while modest size ones
do best in F. So G is a composite scheme that looks at the value of a number and decides which of the other two formats would lead to the most
compact representation, and it then uses that. It is roughly what Java’s
Double.toString method provides, but again we would really like options to indicate precision and field width.
Implement functions which convert Java double values to strings in each of the
above formats.
I might suggest that you start by using toString to do the basic conversion
and then let your code restrict its worry to unpicking that string and re-formatting
the characters. If you decide you want to do the numeric to string conversion from
scratch you should be aware that preserving numeric accuracy is quite hard!
Write a test-suite that compares the strings your code generates with the ones
that String.format produces. Then worry about NaNs, infinities, careful rounding and the like!
Double as bit-patterns
Double.doubleToLongBits takes a double as an argument and returns a
long. The long is the internal IEEE-format bit-pattern that represents the double
The matching function longBitsToDouble accepts a long and manufactures a
double in the same dubious sort of way. Investigate whether there is any double
value x in Java so that
(double)Double.doubleToLongBits(x) == x
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
139
Continued fractions
Any positive number can be expanded as a continued fraction as in
x = x0 +
1
1
x1 + x2 +...
where the values xi are called partial quotients and are all positive integers. If
the original number is rational the continued fraction terminates at some stage.
Otherwise it goes on for ever, and can be viewed as providing an alternative
to the usual decimal expansion of numbers. Instead of writing a value as say
1.414213562. . . the partial quotients would be listed [1, 2, 2, 2, . . . ]. Gosh in fact
for this number it looks as if the continued fraction is astonishingly regular!
The sequence of partial quotients in the expansion of a number are easy to
compute - the first is just obtained by casting the number to an int. The rest can
be obtained from the reciprocal of what you get by subtracting that value from the
original number. Write code to do this and tabulate the first dozen partial quotients
you get in the expansions of the following numbers:
√
3
√
( 5 + 1)/2
√
7
e = 2.71828 . . .
π
4.8 Important features of the class libraries
The coverage thus far has shown the use of some small parts of the Java libraries,
but has also missed a great deal out. In this course I will not have anything like
enough time to describe everything that is available. However there are a few
bits of functionality that either seem to be generally useful enough or sufficiently
fun to be worth covering. The little bits of explanation given here are thus to be
viewed as a sampler of what Java can do for you. If you can work through all these
demonstrations and navigate the documentation of the classes that they introduce
you should have got a reasonably broad idea of the system, and in looking up the
documentation details while working on these cases you will as a side-effect be
noticing what other classes are present. I will only give the most basic possible
demonstrations of the things illustrated here. Full competent use of them can only
come with serious work on rather larger bodies of code. I will also totally ignore
140
CHAPTER 4. BASIC USE OF JAVA
several of the newer parts of the Java class libraries, or to be more precise, I will
leave fuller details of some of these facilities and of the other ones to next year’s
“Concurrent Systems and Applications” course and/or your own private study.
4.8.1 File input and output
The character input and output shown so far has used the pre-defined “standard”
streams System.in and System.out. Obviously in many real applications it is
necessary to access named files.
In many programming languages there is a logical (and practical) distinction
made between files that will contain text and those while will be used to hold
binary information. Files processed as binary are thought of as sequences of 8bit bytes, while ones containing text model sequences of characters. In Java this
distinction has two main manifestations, one of which is somewhat frivolous but
can matter on an every-day basis in the UK while the other is of wider importance
but will not impinge on immediate coursework:
1. Windows and some internet protocols use a pair of characters, carriagereturn and line-feed, to mark the end of a line. Unix and Linux use a single
character (newline). In text mode Java makes whatever adjustments are
needed so that external files adhere to platform conventions, but your Java
code sees just the character ’\n’.
2. In many parts of the world (and in particular in the Far East) text documents
need to cope with alphabets that involve many thousands of symbols. Unicode is designed to be able to cope with these, but there can be a variety
of ways of encoding text as streams of bytes. When working in such an
environment Java can be configures so it knows how to pack and unpack
Unicode using various of the major encoding conventions. But obviously it
will only even try to do this when it knows that the programmer wants data
to be viewed as character-based rather than binary.
Java uses names typically based on the word Stream for binary access, and
Reader and Writer for text. So when you read the documentation expect to find
two broadly parallel sets of mechanisms, one for each style of access!
Java input and output can seem clumsy to start with because almost all of
the functions involved are able to throw exceptions, and it is expected that Java
code using them should be prepared to handle these. This is in fact good because
experience with earlier languages indicates that most programmers do not find
it easy or natural to put error-checks after simple I/O operations, even though
logically almost any of them could fail. For instance writing a single character
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
141
to a file could unexpectedly fail if the disc on which the file lived became full51,
or it it was on a floppy disc and the disc was removed from the drive or had a
scratch, or if there was a hardware glitch in a disc interface. Different but equally
delicate issues arise with output that goes directly to a printer, across a network to
a remote file-system, or with input from a semi-reliable device such as a bar-code
scanner. The Java use of exceptions encourages all programmers to consider I/O
failure right from the start.
There is one final complication about Java input and output that ought to be
mentioned up front. One use of Java is in applets to be embedded within web
pages, and hence sometimes fetched from remote web-sites. It could be bad if
code from an untrusted site could read and write all your files! So Java introduces
the idea of a security manager and can classify code as either trusted or untrusted.
Untrusted code will not be permitted to access the local filing system. The short
form way around this is to make everything you do an application not an applet:
security restrictions are then (by default) not imposed. If you do need to make
applets that access disc or do other things that default controls lock out you need
to impose security on an application then you will need to find out about the
creation of custom Security Policies and signed Java code. I will not describe that
here.
Java provides a rich and somewhat elaborate set of capabilities, but perhaps
a good place to start will be simple reading of text files. The class FileReader
does almost everything you are liable to need: here is a minimal demonstration
that shows that you can use a method called read to read characters, and that it
returns the integer -1 at end of file.
import java.io.*;
public class ReadDemo
{
public static void main(String [] args)
throws IOException
{
Reader r = new FileReader("filename");
int c;
while ((c = r.read()) != -1)
{
System.out.printf(
"Char code %x is \"%<c\"%n", c);
}
r.close();
}
}
51 Or
the user’s quota expired.
142
CHAPTER 4. BASIC USE OF JAVA
There are a significant number of things about this small bit of sample code
that deserve further explanation, and by trying to be minimal the code is not really
very good: an improved version is given soon.
Firstly note that FileReader is in the java.io package so we have an import statement to make use if it easy. Next observe that almost all input and output
functions can raise exceptions, and this code just admits defeat and notes that its
main method might therefor fail. I view it as bad style to do this and strongly
believe that exceptions should be handled more locally.
Now FileReader is a subclass of Reader, which is the general class that
reads from character streams. So I create a FileReader using a constructor that
takes a file-name as its argument but store what I get as just a Reader . This
helps stress to me and remind me that the rest of my code would be equally valid
if using some other sort of Reader, such as one that gets its input from a pipe,
from a string, from characters packed in an array, from a network connection,
by running a character decoder on a stream of bytes or otherwise. The way I do
things here supposes that the data in the file concerned is encoded in the standard
local character-set that Java has been set up for. For reading files imported from
elsewhere in the world you have to do things a more complicated way!
The read method hands back either the numeric code for a character, or the
value -1 to denote end-of-file. It perhaps seems odd that it returns an int not a
char, but doing so allows it to hand back -1 which does not stand for any normal
character. You can of course case the int to a char any time you want to!
After having read the file you are expected to call the close method. If you
fail to do this for an input file you may just leave some machine resources cluttered
and unless you try to open and read very many files without closing any of them
you will probably not feel any pain. However for output files it may sometimes be
that the last bit of your data is not actually sent to the file until you do the close.
You should get into the habit of ensuring that every file you open does get closed.
A much improved version of the same code can be arrived at by handling
the possible exceptions. You may note that FileNotFoundException is a subclass of IOException which is why the throws clause above was sufficient, but
which also allows us to see how the improved code is more precise. When you get
an exception out of Java it can often be useful to print it, in that it is liable to carry
some text that explains further what went wrong. I use finally to guarantee that
the close method of the Reader will always be invoked.
import java.io.*;
public class BetterReadDemo
{
public static void main(String [] args)
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
143
{
Reader r;
try
{
r = new FileReader("filename");
}
catch (FileNotFoundException e)
{
System.out.printf(e);
return;
}
int c;
try
{
while ((c = r.read()) != -1)
{
System.out.printf(
"Char code %x is \"%<c\"%n", c);
}
}
catch (IOException e)
{
System.out.printf("Reading failed (%s)%n", e);
return;
}
finally
{
r.close();
}
}
}
Output to a file is somewhat similar, and if you only ever want to write individual characters and simple strings then FileWriter will suffice. However you
may like to be able to use println and printf when writing data to your file,
and they come in a class called PrintWriter. Unlike the class FileWriter,
PrintWriter hides all exceptions so you do not need to catch them, but you can
check for error using the checkError method and you still need to ensure that
close is called.
import java.io.*;
public class PrintDemo
{
public static void main(String [] args)
{
try
{
PrintWriter w = new PrintWriter("filename");
try
CHAPTER 4. BASIC USE OF JAVA
144
{
w.printf("Hoorah%n");
assert !w.checkError();
}
finally
{
w.close();
}
}
catch (FileNotFoundException e)
{
System.out.println("Sorry!");
}
}
}
This time you must make w a PrintWriter and not just a Writer to gain
access to printf and so on.
If errors arise on a PrintWriter the flag marking them persists so you do not
need to use checkError after every single print statement – every so often and
once when you have generated all that you want to end up in the file will suffice.
Although I have used assert here I probably feel that error checking should be
done always an that something along the lines of
if (w.checkError())
throw new IOException("failure on PrintWriter");
might well be better policy.
The long-winded but more flexible way to access files is to start by creating
an instance of java.io.File. An object of this type can be created using either
a constructor that takes a single String that names the file (as a complete path,
if necessary), or with a two-argument constructor where one argument specifies
the directory to look in and the other the file-name within that directory. A File
object supports methods exists, canRead and canWrite and also one called
isFile, which test for a “normal” file, ie one that is not a directory or any of the
exotic things that in Unix masquerade as sort-of-files. You can pass a File rather
than a string when opening a FileReader or FileWriter.
Other methods available via the File class include ones to check the length
of a file52, rename it, create new directories, list all the files in a directory and
delete files. You can also create a file by giving just a local name (eg such as
"java.tex") and call getAbsolutePath to obtain a fully-rooted file-path that
identifies it. The exact result you get will clearly be system-dependent, and on
one computer I tried that I got back
52 The length reported is liable to count in bytes,
reported differs from system to system.
and so for text files it can well be that the length
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
145
"e:\UNIV\notes\java\java.tex"
while on another it was
"/home/acn1/javanotes/java.tex".
The fact that all these facilities are so conveniently supported may make Java
one of the more useful programming languages for writing file-management utilities. Once again if you look at Java code and compare it against other languages
for very tiny tasks and where previously you would have missed out all error
handling Java can look clumsy — but when you look at more realistic and wellengineered examples it starts to feel much nicer.
Binary data access are useful for cases when your data really is raw data and
not composed of characters. There classes called java.io.FileInputStream
and (of course) FileOutputStream that take a File or a string as an argument
and create streams. They of course throw exceptions if the files can not be opened
as requested. Once a file has been opened you should in due course call the relevant close method to tidy up.
Earlier examples have shown an extra layer of Java constructor arranging to
buffer input in the expectation that that may speed it up. I have done that here too.
Putting these together we might arrive at something like this to copy a file in
binary mode:
String fromName = "source.file";
String toName = "destination.file";
File fromFile = new File(fromName),
toFile = new File(toName);
if (!fromFile.exists() ||
!fromFile.canRead() ||
toFile.exists() ||
!toFile.canWrite())
{
System.out.println("Can not copy");
return;
}
InputStream fromStream =
new BufferedInputStream(
new FileInputStream(fromFile));
try
{
OutputStream toStream =
new BufferedOutputStream(
FileOutputStream(toFile));
try
{
for (;;)
{
toStream.write(fromStream.read());
CHAPTER 4. BASIC USE OF JAVA
146
}
}
catch (EOFException e)
{} // Use exception to break out of for loop
finally
{
toStream.close();
}
}
catch (IOException e)
{
System.out.printf("IO error " + e);
}
finally
{
fromStream.close();
}
This code is in fact not yet complete! It needs yet more try blocks to guard against
FileNotFoundException cases where the two streams are created. But it illustrates how the EOFException can be used to stop processing at end of file, and
demonstrates very clearly that in real file-processing applications most of what
you write will be to do with setting everything up and arranging to handle exceptions, while the central interesting bit of the code may be as short as just
for (;;)
{
toStream.write(fromStream.read());
}
Overall it may seem pretty grim, but in large programs the complication will
still remain at the level of the dozen or so lines shown above, rather than growing
out of control. It is also probable that the visible pain is because writing high quality file-manipulation code is in fact nothing like as easy as earlier programming
languages have tried to make it out to be!
There is a potential down-side in Java being so very insistent that you catch
all these errors, in that it can encourage a style of cop-out that just wraps all your
code in
try
{
...
}
catch (Exception e)
{}
where the block is set up so it catches all sorts of Exception not just the very
special ones that you know are liable to arise, and rather than doing anything it just
ignores the error. This very much defeats the purpose Java is trying to achieve! If
you are (quite reasonably!) in a rush some time at least go:
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
147
try
{
...
}
catch (Exception e)
{
System.out.printf("Exception: %s%n", e);
System.exit(1);
}
so that the exceptions you catch are reported and make your program stop.
The above example used BufferedInputStream which should not have
any effect at all on what your program actually does, but may have an impact
on performance when you work with big files. For binary data there are more
interesting classes that you could use just as easily: ones to compress and and decompress data, encryption and checksumming capabilities. For text data you can
use LineNumberReader in place of BufferedReader and it will keep track of
which line you are on in your input. See the classes FilterInputStream and
FilterReader in the documentation for further details.
4.8.2 Big integers
The Discrete Mathematics course had an extended section where it discussed
highest common factors, modular arithmetic and eventually the RSA encryption
scheme. To refine your understanding of all that you could quite properly want
to code it all up. To make any pretence at all of reasonable security this means
that you need to do a lot of arithmetic on integers that are between 1024 and 1536
bits long. This sort of range of values is about what is required because there is
a serious possibility that numbers smaller than that might be factorisable by the
best current algorithms and fastest current computers. An implementation of RSA
will also need to generate a couple of primes, each with around half that number
of bits.
Java has thought of that and it provides a class java.math.BigInteger
which does essentially everything you could need! And note that printf lets
you print these big values easily.
In this class there are half a dozen constructors. The more obvious ones construct big integers from strings or byte arrays, and a valueOf method allows you
to create a big integer from a long. The two interesting constructors create random big numbers. They both accept an argument that is an object from the class
Random which actually gives them their randomness. One creates an arbitrary
n-bit number while the other creates an n-bit number which is (almost certainly)
prime. For the second of these it is possible to tune the degree of certainty that a
prime has indeed been found by giving a “certainty” argument that tells the con-
148
CHAPTER 4. BASIC USE OF JAVA
structor how hard to work to check things. I might suggest that a value of 50
would be sufficient for all reasonable purposes.
I should provide a rather heavy health warning here. If you use the Javaprovided Random class to help you create private keys or other values of cryptographic significance you will be throwing away almost all the security that the
RSA method could give you, since this random number generator comes too close
to having a predictable behaviour. Specifically there is a chance that to arrange
to get the same “random” values that you do it may suffice for somebody to run
a similar Java program having reset their computer so that their run appears to
happen at the exact time of day that yours did. This may be hard but is nothing
like as hard as factorising 1536-bit integers. If you ever wanted to use serious
encryption you must instead use java.security.SecureRandom. Anybody
really serious about security would think at length before trusting even the things
in java.security: how is it possible to tell that they do what they are supposed
to and that they do not include secret weaknesses? And even if they are honest it
is astonishingly easy to lose all the security you thought you had by some apparently minor clumsiness in how you use your cryptographic primitives. A course
on security later in the Tripos gives much more information about all of this!
Note that some of the functionality in the Java security and encryption packages may be missing or limited unless your installation has provided some level
of assertion that you are not a national of a country that the US Government does
not like and that you are not a terrorist. But as is the way of any such attempt
at blocking access to technology, there are easy to find drop-in replacements not
hampered by (so many) export license issues. You might still be aware that good
encryption is viewed by some as something with significant security implications
and that it should not be given any opportunity to cross international borders until
you at the very very least know what all the rules are! Java itself provides ways
that those who satisfy the correct eligibility conditions can use the standard Java
libraries and obtain unlimited security, via the installation of special “JCE policy
files”.
Once you have made suitable objects of class BigInteger the library provides you with methods to add, subtract, multiply and divide them, to even raise
one big number to a huge power modulo another number (what a give-away about
the expected use of this class!). The function that computes what the Discrete
Mathematics notes called a Highest Common Factor is here known as a Greatest Common Divisor (gcd), but the change of name does not hide any change of
behaviour53.
53 The
java_
security package provides easy to use functions for generating keys and computing message digests and digital signatures. There is a standard extension to Java that supplies
encryption and further functionality: this part may be subject to export regulation and has to be
fetched and installed as a separate item from the main SDK.
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
149
import java.math.*;
import java.util.*;
...
Random r = new Random(); // inadequate!
// use the SecureRandom class instead!!!
// Create two big primes p and q
BigInteger p =
new BigInteger(768, // length in bits
50, // only 1 in 2ˆ50 prob of non-prime
r); // random number source
BigInteger q = new BigInteger(768, 50, r);
// form their product, n, which can be public
BigInteger n = p.multiply(q);
// compute phi = (p-1)*(q-1)
BigInteger bigOne = BigInteger.ONE;
BigInteger pMinusOne = p.subtract(bigOne);
BigInteger qMinusOne = q.subtract(bigOne);
BigInteger phi = pMinusOne.multiply(qMinusOne);
// select a random exponent whose HCF with phi
// is 1.
BigInteger e;
do
{
e = new BigInteger(1536, r);
} while (!phi.gcd(e).equals(bigOne));
// now (n, e) is the public key
...
// Set up a message to encrypt
BigInteger msg =
new BigInteger("12341234"); // silly message
// Encrypt with public key. One line of code!
BigInteger k = msg.modPow(e, n);
...
The code is clearly about as short as it possibly could be. Again let me warn you
that cryptographically satisfactory random number generators are hard to come
by, and that such issues as managing the secure distribution of public keys and
keeping private ones truly private mean that security involves very much more
than just these few lines of code. But Java is clearly making it easy to make a start
on it.
How do you know that the Authorities and not bugging your computer while
you run the above code? How do you know that no traces of the secret information
remain anywhere when you have logged off or even powered down the computer?
The Computer Lab’s security group has a fine track-record of demonstrating that
CHAPTER 4. BASIC USE OF JAVA
150
even apparently safe computing habits leak information to a sufficiently skilful
and ingenious snooper.
4.8.3 Collections
Java has an interface called Collection and a whole range of interesting classes
derived form it. The general idea is that Collection covers ideas like “set”,
“list” and “vector”. In some cases the elements in a collection can be ordered
(in which case the objects included must all implement the Comparable interface54), but might not be. Collections may be implemented as linked lists or as
vectors, but the library classes arrange that when a vector is used it will be enlarged as necessary so that the user does not have to specify a limit to the size of
the collection too early. One sub-case of a Collection is a Map, which provides
for general ways to organize various sorts of table or dictionary. I am not providing any sample code using Collections in this little section since I believe
that when you browse the documentation you will find them easy to cope with.
However it may make sense for me to list more of the names of classes worth
looking at: Collection, Collections, Set, HashSet, TreeSet, Vector,
LinkedList.
The collection classes are keyed to the Java for statement to make it trivial to
iterate over the values in a collection: as has been seen in various of the sample
programs here.
Very typically when you create an instance of a Collection Class you will use
the generics capability to indicate the type of the objects you will keep in it, eg
Vector<String> v = new Vector<String>();
and if you do so Java will know that the values you extract will be of the type
indicated.
4.8.4 Simple use of Threads
A thread is a stream of computation that can go on in parallel with others. The
term is used when the activities are part of a single program, and where there is
no need for security barriers to protect one thread from the next. The more general term used when the extra overhead of protection is needed is process. Java
is one of the first languages to make a big point of having threads supported as
54 An
especially interesting issue here is the way that Java can compare strings. To support the
needs of different nationalities a class Collator is provided, and methods in it can order strings
based on proper Locale-specific understandings of where accented or non-English letters should
go. Alphabetic ordering with international texts is a more complicated business that almost all of
you would have imagined.
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
151
a standard facility. Many systems in the past have had threads, but usually in
rather non-portable form. Almost any program that has to implement a complicated windowed user interface or which accesses remote computers55 will need to
use threads so that one high priority thread can ensure that the user always gets
responses to requests, while several lower priority ones keep on with some bigger
calculation. There are very many subtleties in any program that exhibits concurrency. I will not describe these here, and in consequence I expect that people who
try to make substantial use of threads based on just these notes will get themselves
into deep water. There are two typical bad effects that can arise. In one the system
just locks up as a chain of threads each wait for the others to complete some task.
In the other two threads both attempt to update some data structure at around the
same time and their activities interfere, leaving data in a corrupted state. The Java
keyword synchronized is involved in some of the resolutions of these sorts of
problem.
The example here is supposed to do not much more than to alert you to the
possibility of writing multi-threaded Java programs, and to show how easy it is.
I will start by defining a class that will encapsulate the behaviour I want in the
rather silly thread that I will use here:
class Task extends Thread
{
boolean resultShown;
String result;
int identification;
Task(int i)
{
identification = i;
resultShown = false;
}
public void run()
{
try { sleep(20+100*identification % 77); }
catch (InterruptedException e) { return; }
result = String.valueOf(identification);
}
}
The two critical things are that my class extends Thread and that it implements
run. The method run will be to a thread much what main is to a complete
program. In this case I make my thread do something rather minimal. It goes to
sleep for an amount of time that depends on the argument that was passed to its
constructor, and it then sets one of its variables, result, to a string version of
that value. When created my task also sets a flag that I will use later on to record
whether I have picked up its result.
55 Often
a slow business.
152
CHAPTER 4. BASIC USE OF JAVA
To demonstrate use of threads I will create half a dozen instances of the above,
and then wait around until each has finished its work. As I notice each task completing I will pick up its result and display it. When I have done that I will set
the resultShown flag so that I do not display any result twice. I could surely
find a cleverer way of achieving that, but the solution I use here is at least quite
concise. Once all my threads have finished I will let my main program terminate.
I let my top-level class inherit from Thread just because I want to use sleep in
it so that while waiting for the sub-tasks to finish I am mostly idle.
public class Threads extends Thread
{
static final int THREADCOUNT = 6;
//
//
//
//
public static void main(String[] args)
{
Create and start six threads
Task [] spawn = new Task [THREADCOUNT];
for (int i = 0; i<THREADCOUNT; i++)
{
spawn[i] = new Task(i);
spawn[i].start();
}
System.out.println("All running now");
int stillGoing = THREADCOUNT;
Scan looking for terminated threads
while (stillGoing != 0)
{
for (int i=0; i<THREADCOUNT; i++)
{
if (!spawn[i].isAlive() &&
!spawn[i].resultShown)
print result the first time I notice a thread dead
{
System.out.println("Result from " +
i + " = " + spawn[i].result);
spawn[i].resultShown = true;
stillGoing--;
}
}
System.out.println("One scan done");
sleep for 7 milliseconds between scans to avoid waste
try { sleep(7); }
catch (InterruptedException e) { break; }
}
System.out.println("All done");
}
}
Observe that sleep can raise an exception if the sleeping task receives an in-
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
153
terrupt from elsewhere, and I (have to) catch this and quit. The results I obtain
follow, and you can see traces that show the main program scanning round looking for threads that have finished and also you can see that the different threads
terminate in some curious order. Of course a more worthwhile example would put
real computation into each of the threads and their termination would be based on
how long that took rather than on the artificial sleeping I have used here!
java Threads
All running now
One scan done
One scan done
One scan done
Result from 0 =
Result from 4 =
One scan done
Result from 1 =
One scan done
One scan done
Result from 5 =
One scan done
Result from 2 =
One scan done
One scan done
One scan done
Result from 3 =
One scan done
All done
0
4
1
5
2
3
The reason my example is respectably simple and trouble-free is that the threads
only communicate by receiving data when first created and by delivering something back when they have finished. Inter-process communication beyond that can
be astonishingly hard to get right.
4.8.5 Network access
Java really hit the news as a language for animating your own web pages. One
part of doing this is the set of graphical operations that it supports. Another less
instantly visible but equally important thing is the ability to connect to remote
computers and retrieve data from them. The set of rules that make up HTTP56 are
what defines the World Wide Web. Standard Java libraries provide various degrees
of ability to connect using it. The small program shown here links through to a
56 HyperText
Transfer Protocol.
154
CHAPTER 4. BASIC USE OF JAVA
fixed location named as its fire command-line argument and displays the data
found there. This data comes out as an HTML document with lots of tags that are
enclosed in angle brackets.
// Read file from a possibly remote web server
import java.net.*;
import java.io.*;
public class Webget
{
public static void main(String [] args)
{
URL target;
try
{
target = new URL(args[0]);
}
catch (MalformedURLException e)
{ return; } // Badly formed web-page address
try
{
URLConnection link = target.openConnection();
link.connect(); // connect to remote system
// Now just for fun I display size and type information
// about the document that is being fetched. Note that
// documents might be pictures or binary files as well
// as just text!
System.out.println("size = " +
link.getContentLength());
System.out.println("type = " +
link.getContentType());
// getInputStream() gives me a handle on the content, and
// I rather hope it is text. In that case I can get the
// characters that make it up from the InputStream.
Reader in = new InputStreamReader(
link.getInputStream());
int c;
// Crude echo of text from the document to the screen.
// It will have lots of HTML encoding in it, I expect.
while ((c = in.read()) != -1)
System.out.print((char)c);
}
// A handler is needed in case exceptions arise.
catch (IOException e)
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
155
{ System.out.println("IO error on link"); }
// I am lazy here and do not close anything down.
}
}
// end of Webget.java
The data stored on the CL teaching support pages in mid February 1998 started
off as follows, apart from the fact that I have split some of the lines to make the
text fit neatly on the pages of these notes. It has of course changed by now! I keep
this old material in the notes out of nostalgia.
<HTML>
<HEAD>
<TITLE>Comp.Sci. Teaching pages</TITLE>
</HEADER>
<BODY>
<H1> Computer Science teaching material on Thor</H1>
<P>
<UL>
<LI><A HREF="Java/java-index.html"> Some Information
about Java </A> (on this server)
...
The main message here is that accessing a remote web-site is just about as
trivial as reading from a local file.
4.8.6 Menus, scroll bars and dialog boxes
Back when Java 1.2 was released Sun finalised a whole set of windows management code that they called Swing. This extended and in places replaced earlier
windowing capabilities that were known as AWT. I believe that by now it is proper
to use the Swing versions of things even in those cases where older AWT versions
remain available. You can tell that you are doing that when you use a lot of class
names that start with a capital “J”!
The code presented here is called MenuApp and is a pretty minimal demonstration of menus! I will use this example to show how something can be both
an application and an applet. The “application” bit of course defined a method
called main, and this just sets up a window (frame) that can contain the applet
stuff. There is a bit of mumbo jumbo to allow the application to stop properly
CHAPTER 4. BASIC USE OF JAVA
156
when the window is closed. As usual I will show the inner bit of the code first –
the fragment that actually does the work:
public static void main(String[] args)
{
Menuapp window = new Menuapp();
JFrame f = new JFrame("Menu Demonstration");
f.addWindowListener(new WindowAdapter()
{
public void windowClosing(WindowEvent e)
{
System.exit(0);
}
});
f.getContentPane().add(window, BorderLayout.CENTER);
f.setSize(600, 400);
f.setVisible(true);
}
I should point out the syntax
new WindowAdapter()
{
public void windowClosing(WindowEvent e)
{
System.exit(0);
}
}
}
which extends the class WindowAdapter
to produce a new (anonymous) class.
In this new class it overrided the
windowClosing method. It then creates an instance of the new anonymous
class. This arrangement is known as an
“Inner Class” and can be very handly
when you need a small variant on some
existing class and will use it just once
so that giving it a name would be overclumsy.
The constructor for Menuapp will
establish a menu bar at the top of itself, then makes menus on that bar, and
places menu items on each menu. In
much the way that mouse events were
dealt with by registering a handler for
them it is necessary to implement an
Menuapp running
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
157
interface called ActionListener and
tell each menu to report via it. The report hands down an ActionEvent from
which it is possible to extract the name of the menu item (and if need be which
menu it was on) that was activated. I illustrate this by showing how to pop up
a dialog box for selecting a file, although once I have the name of the file I just
display that in the text area rather than actually opening it.
I put a scrollable editable window for the text. The version I use could in fact
support multi-colour text in mixed fonts and with icons and other things interleaved with the words: finding out about that is an exercise for those of you who
feel keen. You will also find that I have coded this using the “swing” facilities
(ie it will not compile on a simple un-extended installation of JDK 1.1.x), and the
arrangements for selecting a file and for making the text window scrollable relate
to that. The inclusion of javax.swing classes gives access to the relevant bits
of the class libraries. Furthermore the code can be run as either an applet or an
application. So lots of things are being illustrated at once. You are not expected
to be able to follow all of them at first, but maybe the code will be a useful model
when sometime later you do need to use some of these facilities in anger. The
complete code follows:
// Demonstration of Menus and a window created
// from an application rather than an applet.
// A C Norman
1998-2000
import
import
import
import
java.awt.*;
java.awt.event.*;
javax.swing.*;
javax.swing.text.*;
public class Menuapp extends JApplet
implements ActionListener
{
// This can be run as either an application or an applet!
public static void main(String[] args)
{
Menuapp window = new Menuapp();
JFrame f = new JFrame("Menu Demonstration");
f.addWindowListener(new WindowAdapter()
{
public void windowClosing(WindowEvent e)
{
System.exit(0);
}
});
f.getContentPane().add(window,
BorderLayout.CENTER);
CHAPTER 4. BASIC USE OF JAVA
158
f.setSize(600, 400);
f.setVisible(true);
}
// All real work happens because of this
// constructor. I create a JTextPane to hold
// input & output and make some menus.
JTextPane text;
Container cc;
public Menuapp()
{
cc = getContentPane();
text = new JTextPane();
text.setEditable(true);
text.setFont(
new Font("MonoSpaced", Font.BOLD, 24));
JScrollPane scroll = new JScrollPane(text,
JScrollPane.VERTICAL_SCROLLBAR_AS_NEEDED,
JScrollPane.HORIZONTAL_SCROLLBAR_NEVER);
cc.add(scroll);
// Menus hang off a menu bar and contain menu items
JMenuBar bar;
JMenu mFile, mEdit, mHelp;
JMenuItem mNew, mOpen, mSave, mCut,
mPaste, mContents;
// Create a menu bar first and add it to the Frame
bar = new JMenuBar();
setJMenuBar(bar);
// Create a menu and add it to the MenuBar
mFile = new JMenu("File");
bar.add(mFile);
// Create menu items and add to menu
mNew = new JMenuItem("New");
mFile.add(mNew);
mOpen = new JMenuItem("Open"); mFile.add(mOpen);
mSave = new JMenuItem("Save"); mFile.add(mSave);
mEdit = new JMenu("Edit");
bar.add(mEdit);
mCut = new JMenuItem("Cut");
mEdit.add(mCut);
mPaste = new JMenuItem("Paste");mEdit.add(mPaste);
mHelp = new JMenu("Help Menu");
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
159
bar.add(mHelp);
mContents = new JMenuItem("Contents");
mHelp.add(mContents);
// Each menu has to be activated to be useful.
mNew.addActionListener(this);
mOpen.addActionListener(this);
mSave.addActionListener(this);
mCut.addActionListener(this);
mPaste.addActionListener(this);
mContents.addActionListener(this);
}
//
//
//
//
//
When a menu item is selected this gets called,
and getActionCommand() retrieves the text from
the menuItem. Here I clear the area when New
is used, and do something with Open, but otherwise
just display a message.
public void actionPerformed(ActionEvent e)
{
String action = e.getActionCommand();
try
{
if (action.equals("New")) text.setText("");
else if (action.equals("Open")) openFile();
else
{
StyledDocument s =
text.getStyledDocument();
s.insertString(s.getLength(),
"Request was " + action + "\n", null);
}
}
catch (BadLocationException e1)
{}
}
void openFile() throws BadLocationException
{
JFileChooser d = new JFileChooser("Open a file");
if (d.showOpenDialog(cc) ==
JFileChooser.APPROVE_OPTION)
{
StyledDocument s = text.getStyledDocument();
s.insertString(s.getLength(),
"Load file \"" +
d.getSelectedFile().getAbsolutePath() +
CHAPTER 4. BASIC USE OF JAVA
160
"\"\n", null);
}
}
}
// end of Menuapp.java
You should expect that extending the above example or writing your own code
that sets up controllable visual effects will cause you to have to do rather a lot
of reading of the class library documentation to plan which classes you will derive items from. Also when you have mastered the basics of GUI construction by
hand you will very probably want to take advantage of one of the Java development environments that can set up frameworks for user-interfaces for you in really
convenient ways.
4.8.7 Exercises
Replacement for “ls”
On Unix the command ls lists all the files in the current directory. With a
command-line flag -R it also lists members of sub-directories. Investigate the
Java File class and see how much of the behaviour of ls (or the DOS/Windows
dir) you can re-create.
RSA
The code fragment above suggests how to create a public key and then how to use
it to encrypt a message once that message has been reduced to a BigInteger
of suitable size. Flesh this out with code that can use the private key to decrypt
messages, and with some scheme that can read text from the standard input (or a
file, maybe), split it up into chunks and represent each chunk as a BigInteger.
You might also want to investigate the Java Cryptographic Architecture and
find out what is involved in creating cryptographic-grade random values. You
should be made very aware that the ordinary Java random number generator does
not pretend that the values it returns to you are suitable for use in security applications.
Then do a literature search to discover just what you are permitted to do with
an implementation of an idea that has been patented57 and also what the Secu57 The
main RSA US patent expired on the 20th September 2000, but that does not necessarily mean that all associated techniques and uses are unprotected. Also note that there are other
public key methods for both digital signatures and for encryption where the original patents have
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
161
rity Policies of various countries are about the use, import and export of strong
encryption technology.
Note again that Java provides the specification of a security library that would
do all this for you if it were not for the USA government’s export restrictions, and
if these restrictions do not apply to you you could use the Java strong security.
There are international re-implementations of this library that can be fetched and
used here. See for instance www.openjce.org. But also be aware that exporting
code that includes strong encryption may be subject to government restriction.
Big Decimals
The class BigDecimal builds on BigInteger to provide for long precision decimal fractions. When a BigDecimal is created it will behave as if it has some
specified number of digits to the right of its decimal point, but as arithmetic is
carried out there can be any number of digits generated before the decimal point.
For any number z one can define a sequence of values xi by x0 = 1 and
√
xi+1 = (xi + z/xi )/2. This sequence will converge to z, and once it gets even
reasonably close it homes in rapidly, roughly doubling the number of correct significant values at each step. For finding square roots of numbers between 0.5 and
2 (say) the starting value x0 = 1 will be quite tolerable.
If one wanted the square root of a number larger than 2 or smaller than 0.5
it
would
make sense to augment the recurrence formula by use of the identity
√
√
4z = 2 z to pre-normalise z.
Implement the method and use it to create a table of square roots of the integers
from 1 to 10 all to 100 decimal places.
If you can perform high-precision arithmetic on floating point numbers you
should try the following calculation. I am going to use informal syntax with
double constants written where you will need to use BigDecimals, and I will
use the ordinary operators to work on values not calls to suitable methods in the
BigDecimal class. I am also not going to tell you here either how many cycles
of the loop you should expect the code to obey or what the values it prints out will
look like! But I suggest that you work to (say) 1000 decimals and see what you
get.
a = 1.0;
b = 1/sqrt(2.0);
u = 0.25;
x = 1.0;
pn = 4.0;
do
also expired, and so using these can also be of interest without patent worries about deployment.
Investigate “El Gamal”.
CHAPTER 4. BASIC USE OF JAVA
162
{
p = pn;
y = a;
a = (a+b)/2.0;
b = sqrt(y*b);
u = u-x*(a-y)*(a-y);
x = 2*x;
pn = a*a/u;
print(pn);
} while (pn < p);
When you have investigated this for yourself try looking for arithmetic-geometric
mean (and the cross-references from there!) in Hakmem.
Mandelbrot set (again)
Adjust the Mandelbrot set program, as considered in an earlier exercise, so that at
the start (maybe in an init method?) it allocates a 400 by 400 array of integers
all initially set to 0. Change the paint method so that it just displays colours as
recorded in this array, and have a separate thread that (a) computes values to go in
the array and (b) uses repaint at the end of each of its resolution cycles to bring
make sure that the screen is brought fully up to date then. The objective here is to
arrange that each pixel is only calculated once and that thus the paint method is
a lot quicker.
Further extension of this code would add mouse support so that (as a start) a
mouse click would cause the display to re-centre at the selected point and start
drawing again using half the previous range. Those of you who get that far are
liable to be able to design more flexible and convenient user-driven controls than
that, I expect!
Tickable exercise 7
Tick 5 was the largest single exercise. Tick 6 was a fairly easy example of getting
an Applet working. This final exercise is intended to be a reasonably manageable
one to end off the series of Java practicals.
The illustration of network code that I gave represents a minimal way of accessing a remote file.
1. Extend it so that it can be used as
java Webget URL destination
to fetch the contents of an arbitrary web document and store it on the local
disc. If an explicit destination file is not specified your program should
display the document fetched on the screen;
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
163
2. Investigate document types and encodings and try to arrange that text documents are fetched as streams of characters while binary documents come
across the net verbatim;
3. Optional: Some files on the web are of type “zip”, being compressed.
Java provides a set of library functions for uncompressing such files. Use
it so that when a zipped file is fetched the data that arrives locally is the
decompressed form of it.
4. Very optional: A javax.swing.JEditorPane can be set up so that it
renders (a subset of) HTML for you. Using this perhaps you could go a
significant way towards implementing your own light-weight web browser
and you might only need a fairly modest amount of Java!
(End of tickable exercise)
Yet further network-related code
The parts of the Java libraries that
can fetch material from remote machines understand quite well that you
will only occasionally be fetching
raw text, and that fairly often the data
to be accessed will be a picture or
a sound. Investigate the documentation and sort out how to use these
facilities. An image from a local
student-run web site is shown here to
illustrate the usefulness of these facilities. Further parts of the network
classes allow you to detect (to some
extent) what sort of data is about to
be fetched from a web location so
that different sorts of data can each
be handled in the most appropriate
manner.
If you could write code that located words enclosed in angle brackets within text, and lay text out in a Figure 4.8: www.arthurnorman.org!
window in such a way that you could
tell which word was the subject of a mouse click you might find yourself half-way
to writing your very own web browser!
CHAPTER 4. BASIC USE OF JAVA
164
A minimal text editor
The menu example already allows you to re-position the cursor and type in extra
text. I have shown how to identify files to be loaded and saved. The TextArea
class provides methods that would implement cut and paste. Put all these together
to get a notepad-style editor.
More Fonts
The Unicode example showed that it is easy to select which font Java uses to
paint characters. The class FontMetrics then provides methods that allow you
to obtain very detailed measurements of both individual characters and rows of
them. Using all this you can get rather fine control over visual appearance. Using
however many of few of these facilities you like create a Java applet that displays
the word LATEX in something like the curious layout that the designers of that
text-formatting program like to see.
Two flasks
This was a puzzle that I found in one of the huge anthologies of mathematical
oddities:
An apothecary has two spherical flasks, which between them hold exactly 9 measures of fluid. He explains this to his friend the mathemagician, and adds that the glass-blower who makes his flasks can make
them perfectly spherical but will only make flasks whose diameter is
an exact rational number. The mathemagician looks unimpressed and
says that the two flasks obviously have diameters 1 and 2, so their
volume is proportional to 13 + 23 = 9. Hmmm says the apothecary,
that would have worked, but I already had a pair like that. This pair of
flasks has exactly the same total capacity, still has linear dimensions
that are exact fractions but the neither flask has diameter 1.
Find a size that the smaller of his two flasks might have had.
This is clearly asking for positive solutions to x3 + y3 = 9 with the solution
being a neat fraction. In an earlier exercise you were invited to write a class to
handle fractions. Before continuing with this one you might like to ensure that
you have a version of it that uses BigInteger just to be certain that overflow
will not scupper you.
Here is an attack you can make on the problem. We are interested in solutions
3
to x + y3 = 9 and know an initial solution x = 1, y = 2. Imagine the cubic equation
drawn as a graph (I know it is an implicit equation, so for now I will be happy if
4.8. IMPORTANT FEATURES OF THE CLASS LIBRARIES
165
you imagine that there is a graph and do not worry about actually drawing it!). The
point (1,2) lies on the curve. Draw the tangent to the curve at (1,2). This straight
line will have an equation of the form lx + my + n = 0 for some l, m and n which
you can find after you have done a little differentiation. Now consider where the
straight line meets the curve. If one eliminated y between the two equations the
result is a cubic in x, but we know that x = 1 is one solution (because the curve
and line meet there). In fact because the line is a tangent to the curve that is a
double solution, and thus the cubic we had will necessarily have a factor (x − 1)2.
If you divide that out you get a linear equation that gives you the x co-ordinate of
the other place where the tangent crosses the curve. Solve that and substitute into
the line equation to find the corresponding value of y. The pair x, y were found by
solving what were in the end only linear equations, and so the numbers must be at
worst fractions. This has therefore given another rational point on the cubic curve.
What you get there is not a solution to the problem as posed, since one of x and
y will end up negative. But maybe if you take a tangent at the new point that will
lead you to a third possibility and perhaps following this path you will eventually
manage to find a solution that is both rational and where both components are
positive.
Write code to try this out, and produce the tidiest fractional solution to the
original puzzle that you can, ie the one with smallest denominators.
166
CHAPTER 4. BASIC USE OF JAVA
Figure 4.9: Odds and Ends follow: watch what the lecturer happens to cover.
Chapter 5
Designing and testing programs in
Java
At this stage in the “Programming in Java” course I will start to concentrate more
on the “programming” part than the “Java”. This is on the basis that you have now
seen most of the Java syntax and a representative sampling of the libraries: now
the biggest challenge is how to use the language with confidence and competence.
In parallel with this course you get other ones on Software Engineering. These
present several major issues. One is that errors in software can have serious consequences, up to and including loss of life and the collapse of businesses. Another is
that the construction of large computer-related products will involve teams of programmers working to build and support software over many years, and this raises
problems not apparent to an individual programmer working on a private project.
A third is that formal methods can be used when defining an planning a project to
build a stable base for it to grow on, and this can be a great help. The emphasis is
on programming in the large, which is what the term “Software Engineering” is
usually taken to have at its core. Overall the emphasis is on recognition of the full
life-time costs associated with software and the management strategies that might
keep these costs under control.
The remaining section of this Java course complements that perspective and
looks at the job of one person or a rather small group, working on what may well
be a software component or a medium sized program rather than on a mega-scale
product. The intent of this part of the course is to collect together and emphasise
some of the issues that lie behind the skill of programming. Good programmers
will probably use many of the techniques mentioned here without being especially
aware of what they are doing, but for everybody (even the most experienced) it can
be very useful to bring these ideas out into the open. It seems clear to me that all
computer science professionals should gain a solid grounding carrying out small
projects before they actually embark on larger ones, even though a vision of what
167
168
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
will be needed in big projects help shape attitudes and habits.
It is at least a myth current in the Computer Laboratory that those who intend
to become (mechanical) engineers have as an early part of their training an exercise where they fashion a simple cube of solid metal, and they are judged on the
accuracy of their work. Such an activity can be part of bringing them into direct
touch with the materials that they will later be working with in more elaborate
contexts. It gets them to build experience with tools and techniques (including
those of measurement and assessment). Programming in the small can serve similar important purposes for those who will go on to develop large software systems:
even when large projects introduce extra levels of complication and risk the design
issues discussed here remain vital.
One of the major views I would like to bring to the art (or craft, or science, or
engineering discipline, depending on how one likes to look at it) of programming
is an awareness of the value of an idea described by George Orwell in his book
“1984”. This is doublethink, the ability to believe two contradictory ideas without
becoming confused. Of course one of the original pillars of doublethink was the
useful precept Ignorance is Strength, but collections of views specifically about
the process of constructing programs. These notes will not be about the rest of
the association of computers with surveillance, NewSpeak or other efficiencyenhancing ideas. The potentially conflicting views about programming that I want
to push relate to the prospect of a project succeeding. Try to keep in your minds
both the idea Programming is incredibly difficult and this program will never work
correctly: I am going to have to spend utterly hopeless ages trying to coax it
into passing even the most minimal test cases and its optimistic other face, which
claims cheerfully In a couple of days I can crack the core of this problem, and
then it will only take me another few to finish off all the details. These days even
young children can all write programs. The concise way to express this particular
piece of doublethink (and please remember that you really have to believe both
part, for without the first you will make a total botch of everything, while without
the second you will be too paralysed ever to start actual coding), is
Writing programs is easy.
A rather closely related bit of doublethink alludes both to the joy of achievement when a program appears to partially work and the simultaneous bitter way in
which work with computers persistently fail. Computers show up our imperfections and frailties, which range through unwillingness to read specifications via
inability to think straight all the way to incompetence in mere mechanical typing
skills. The short-hand for the pleasure that comes from discovering one of your
own mistakes, and having spent many frustrating hours tracking down something
that is essentially trivial comes out as
169
Writing programs is fun.
A further thing that will be curious about my presentation is that it does not
present universal and provable absolute truths. It is much more in the style of
collected good advice. Some of this is based on direct experience, other parts has
emerged as an often-unstated collective view of those who work with computers.
There are rather fewer books covering this subject than I might have expected.
There is a very long reading list posted regularly on comp.software-eng,
but most of it clearly assumes that by the time things get down to actually writing
programs the reader will know from experience what to do. Despite the fact that it
is more concerned with team-work rather than individual programming I want to
direct you towards the Mythical Man Month[7], if only for the cover illustration of
the La Brea Tar Pits1 with the vision that programmers can become trapped just as
Figure 5.1: The La Brea tar pits.
the Ice Age mammoths and so on were. Brooks worked for IBM at a time that they
were breaking ground with the ambitious nature of their operating systems. The
analogous Microsoft experience is more up to date and can be found in Writing
1 You
may not be aware that the tar pits are in the middle of a thoroughly built-up part of
Los Angeles, and when visiting them you can try to imagine some of the local school-children
venturing too far and getting bogged down, thus luring their families, out for a week-end picnic,
to a sticky doom.
170
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
Solid Code[18] which gives lots of advice that is clearly good way outside the
context of porting Excel between Windows and the Macintosh. If you read the
Solid Code book you will observe that it is concerned almost entirely with C, and
its examples of horrors are often conditioned by that. You should therefore let it
prompt you into thinking hard about where Java has made life safer than C and
where it has introduced new and different potential pitfalls.
For looking at programming tasks that are fairly algorithmic in style the book
by Dijkstra[10] is both challenging and a landmark. There are places where people have collected together some of the especially neat and clever little programs
they have come across, and many of these indeed contain ideas or lessons that may
be re-cyclable. Such examples have come to be referred to as “pearls”[4][5]. Once
one has worked out what program needs to be written ideas (now so much in the
mainstream that this book is perhaps now out of date) can be found in one of the
big presentations by some of the early proponents of structured programming[19].
Stepping back and looking at the programming process with a view to estimating
programmer productivity and the reliability of the end product, Halstead[14] introduced some interesting sorts of software metrics, which twenty years on are
still not completely free of controversy. All these still leave me feeling that there
is a gap between books that describe the specific detail of how to use one particular
programming language, and those concerned with large scale software engineering and project management. To date this gap has generally been filled by an apprentice system where trainee programmers are invited to work on progressively
larger exercises and their output is graded and commented on by their superiors.
Much like it is done here! With this course I can at least provide some background thoughts that might help the apprentices start on their progression a little
more smoothly.
When I started planning a course covering this material it was not quite obvious how much there was going to be for me to say that avoided recitations of
the blindingly obvious and that was also reasonably generally applicable. As I
started on the notes it became clear that there are actually a lot of points to be
covered. To keep within the number of lectures that I have and to restrict these
notes to a manageable bulk I am therefore restricting myself (mostly) to listing
points for consideration and giving as concrete and explicit recommendations as I
can: I am not including worked examples or lots of anecdotes that illustrate how
badly things can go wrong when people set about tasks in wrong-minded ways.
But perhaps I can suggest that as you read this document you imagine it expanded
into a very much longer presentation with all that additional supporting material
and with a few exercises at the end of each section. You can also think about all
the points that I raise in the context of the various programming exercises that you
are set or other practical work that you are involved with.
5.1. DIFFERENT SORTS OF PROGRAMMING TASKS
171
5.1 Different sorts of programming tasks
Java experience illustrates very clearly that there are several very different sorts of
activity all of which are informally refereed to as “programming”. On style of use
of Java — of great commercial importance — involves understanding all of the
libraries and stringing together uses of them to implement user interfaces and to
move data around. In such cases the focus is entirely on exploiting the libraries,
on human factors and on ensuring that the code’s behaviour agrees with the manual or on-line documentation that its users will work from. At another extreme
come tasks that involve the design of many new data-structures and algorithmic
innovations in their use. Often in this second style of program there will also be
significant concern over efficiency.
Given that there are different sorts of software it might be reasonable to start
with a classification of possible sorts of program. There are three ways in which
this may help with development:
1. Different sorts of computer systems are not all equally easy to build. For
instance industrial experience has shown repeatedly that the construction of
(eg) an operating system is very much harder than building a (again eg) a
compiler even when the initial specifications and the amount of code to be
written seem very similar. Thinking about the category into which your particular problem falls can help you to plan time-scales and predict possible
areas of difficulty;
2. The way you go about a project can depend critically on some very high
level aspects of the task. A fuller list of possibilities is given below, but
two extreme cases might be (a) a software component for inclusion in a
safety-critical part of an aero-space application, where development budget
and timescale are subservient to an over-riding requirement for reliability,
and (b) a small program being written for fun as a first experiment with a
new programming language, where the program will be run just once and
nothing of any real important hands on the results. It would be silly to carry
forward either of the above two tasks using a mind-set tuned to the other:
knowing where one is on the spectrum between can help make the selection
of methodology and tools more rational;
3. I will make a point in these notes that program development is not something to be done in an isolated cell. It involves discussing ideas and progress
with others and becoming aware of relevant prior art. Thinking about the
broad area in which your work lies can help you focus on the resources
worth investigating. Often some of these will not be at all specific to the
172
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
immediate description of what you are suppose to achieve but will concerned with very general areas such as rapid prototyping, formal validation,
real-time responsiveness, user interfaces or whatever.
I will give my list of possible project attributes. These are in general not
mutually exclusive, and in all real cases one will find that these are not yes–no
choices but more like items to be scored from 1 to 10. I would like to think
of them as forming an initial survey that you should conduct before starting any
detailed work on your program just to set it in context. When you find one or
two of these scoring 9 or 10 out of 10 for relevance you know you have identified
something important that ought to influence how you approach the work. If you
find a project scores highly on lots of these items then you might consider trying
to wriggle out of having to take responsibility for it, since there is a significant
chance that it will be a disaster! The list here identifies potential issues, but does
not discuss ways of resolving them: in many cases the project features identified
here will just tell you which of the later sections in these notes are liable to be the
more important ones for your particular piece of work. The wording in each of
my descriptions will be intended to give some flavour of how severe instances of
the issue being discussed can cause trouble, so keep cheerful because usually you
will not be plunging in at the really deep end of the pool.
Ill-defined One of the most common and hardest situations to face up to is when
a computer project is not clearly specified. I am going to take this case to
include ones where there appears to be a detailed and precise specification
document but on close inspection the requirements as written down boil
down to “I don’t know much about computer systems but I know what I
like, so write me a program that I will like, please.” Clearly the first thing to
do in such a case is to schedule a sub-project that has the task of obtaining a
clear and concise description of what is really required, and sometimes this
will of itself be a substantial challenge;
Safety-critical It is presumably obvious that safety-critical applications need exceptional thought and effort put into their validation. But this need for reliability is far from an all-or-nothing one, in that the reputation of a software
house (or indeed the grades obtained by a student) may depend on ensuring
that systems run correctly at least most of the time, and that their failure
modes appear to the user to be reasonable and soft. At the other extreme
it is worth noting that in cases where robustness of code and reliability of
results are not important at all (as can sometimes be the case, despite this
seeming unreasonable) that fact can be exploited to give the whole project a
much lighter structure and sometimes to make everything very much easier
to achieve. A useful question to ask is “Does this program have to work
5.1. DIFFERENT SORTS OF PROGRAMMING TASKS
173
correctly in all circumstances, or does it just need to work in most common
cases, or indeed might it be sufficient to make it work in just one carefully
chosen case?”
Large It is well established that as the size of a programming task increases the
amount of work that goes into it grows much more rapidly than the number
of lines of code (or whatever other simple measurement you like) does. At
various levels of task size it becomes necessary to introduce project teams,
extra layers of formal management and in general to move away from any
pretence that any single individual will have a full understanding of the
whole effort. If your task and the associated time-scales call for a team of
40 programmers and you try it on your own maybe you will have difficulty
finishing it off! Estimating the exact size that a program will end up or just
how long it will take to write is of course very hard, but identifying whether
it can be done by one smart programmer in a month or if it is a big team
project for five years is a much less difficult call to make.
Shifting sands If either project requirements or resources can change while software development is under way then this fact needs to be allowed for. Probable only a tiny minority of real projects will be immune from this sort of distraction, since even for apparently well-specified tasks it is quite usual that
experience with the working version of the program will lead to “wouldn’t
it be nice if . . . ” ideas emerging even in the face of carefully discussed and
thought out early design decisions that the options now requested would not
be supportable. Remember that Shifting Sands easily turn into Tar Pits!
Figure 5.2: The museum frieze at La Brea.
Urgent When are you expected to get this piece of work done? How firm is the
deadline? If time constraints (including the way that this project will compete with other things you are supposed to do) represents a real challenge it
174
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
is best to notice that early on. Note that if, while doing final testing on your
code, you find that it has a bug in it there may be no guarantee that you can
isolate or fix this to any pre-specified time-scale. This is because (at least
for most people!) there is hardly any limit to the subtlety of bugs and the
amount of time and re-work needed to remove them. If the delivery date
for code is going to be rigidly enforced (as is the case with CST final year
projects!) this fact may be important: even if there is plenty of the project
as a whole a rigid deadline can make it suddenly become urgent at the last
minute;
Unrealistic It is quite easy to write a project specification that sounds good, but is
not grounded in the real world. A program that modelled the stock markets
and thereby allowed you to predict how to manage your portfolio, or one to
predict winning numbers in the national lottery, or one to play world-class
chess. . . Now of course there are programs that play chess pretty well, and
lots of people have made money out of the other two projects (in one case
the statistics that one might apply is much easier than the other!), but the
desirability of the finished program can astonishingly often blind one to the
difficulties that would arise in trying to achieve it. In some cases a project
might be achievable in principle but is beyond either today’s technology or
this week’s budget, while in other cases the idea being considered might not
even realistic given unlimited budget and time-scales. There are of course
places where near-unreasonable big ideas can have a very valuable part to
play: in a research laboratory a vision of one of these (currently) unrealistic goals can provide direction to the various smaller and better contained
projects that each take tiny steps towards the ideal. At present my favourite
example of something like this is the idea of nanotechnology with armies of
molecular-scale robots working together to build their own heirs and successors. The standard example of a real project that many (most?) realistic
observers judged to be utterly infeasible was the “Star Wars” Strategic Defence Initiative, but note that at that sort of level the political impact of
even starting a project may be at least as important as delivery of a working
product!
Multi-platform It is a luxury if a program only has to work on a single fixed computer system. Especially as projects become larger there is substantial extra
effort required to keep them able to work smoothly on many different systems. This problem can show up with simple issues such as word-lengths,
byte-arrangements in memory and compiler eccentricities, but it gets much
worse as one looks at windowed user interfaces, multi-media functions, network drivers and support for special extra plug-in hardware;
5.1. DIFFERENT SORTS OF PROGRAMMING TASKS
175
Long life-time The easiest sort of program gets written one afternoon and is
thrown away the next day. It does not carry any serious long-term support
concerns with it. However other programs (sometimes still initially written
in little more than an afternoon) end up becoming part of your life and get
themselves worked and re-worked every few years. In my case the program
I have that has the longest history was written in around 1972 in Fortran,
based on me having seen one of that year’s Diploma dissertations and having (partly unreasonably) convinced myself I could do better. The code was
developed on Titan, the then University main machine. I took it to the USA
with me when I spent a year there and tried to remove the last few bugs and
make it look nicer. When the University moved up to an IBM mainframe I
ran it on that, and at a much later stage I translated it (semi automatically)
into BBC basic and ran it (very slowly) on a BBC micro. By last year I had
the code in C with significant parts of the middle of it totally re-written, but
with still those last few bugs to find ways of working around. If I had been
able to predict when I started how long this would be of interest to me for
maybe I would have tried harder to get it right first time! Note the radical
changes in available hardware and sensible programming language over the
lifetime of this program;
User interface For programs like modern word processors there is a real chance
that almost all of the effort and a very large proportion of the code will go
into supporting the fancy user interface, and trying to make it as helpful
and intuitive as possible. Actually storing and editing the text could well
be fairly straight forward. When the smoothness of a user interface is a
serious priority for a project then the challenge of defining exactly what
must happen is almost certainly severe too, and in significant projects will
involve putting test users in special usability laboratories where their eyemovement can be tracked by cameras and their key-strokes can be timed.
The fact that an interface provides lots of windows and pull-down menus
does not automatically make it easy to use;
Diverse users Many commercial applications need to satisfy multiple users with
diverse needs as part of a single coherent system. This can extend to new
computer systems that need to interwork seamlessly with multiple existing
operational procedures, including existing computer packages. Some users
may be nervous of the new technology, while others may find excessive
explanation an offensive waste of their time. The larger the number of interfaces needed and the wider the range of expectations the harder it will be
to make a complete system deliver total satisfaction;
Speed critical Increasingly these days it makes sense to buy a faster computer if
176
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
Figure 5.3: The search for speed can lead to eccentric-looking results.
some task seems to take a little longer than is comfortable. However there
remain some areas where absolute performance is a serious issue and where
getting the very best out of fixed hardware resources can give a competitive
edge. The case most in my mind at present is that of (high security) encryption, where the calculations needed are fairly demanding but where there
is real interest in keeping some control over the extra hardware costs that
user are expected to incur. If speed requirements lead to need for significant assembly code programming (or almost equivalently to the design of
task-specific silicon) then the resource requirements of a project can jump
5.1. DIFFERENT SORTS OF PROGRAMMING TASKS
177
dramatically. If in the other hand speed is of no importance at all for some
task it may become possible to use a higher level programming system,
simpler data structures and algorithms and generally save a huge amount of
aggravation;
Real time Real-time responsiveness is characteristic of many control applications. It demands that certain external events be given a response within
a pre-specified interval. At first this sounds like a variant on tasks that are
just speed-critical, but the fine granularity at which performance is specified tends to influence the entire shape of software projects and rule out
some otherwise sensible approaches. Some multi-media applications and
video games will score highly in this category, as will engine management
software for cars and flight control software for airports;
Memory critical A programming task can be made much harder if you are tight
on memory. The very idea of being memory-limited can feel silly when we
all know that it is easy to go out and buy another 64 Mbytes for (currently)
of the order of £502 . But the computer in your cell-phone will have an
amount of memory selected on the basis of a painful compromise between
cost (measured in pennies), the power drain that the circuitry puts on the
battery (and hence the expected battery life) and the set of features that can
be supported. And the software developers are probably give the memory
budget as a fixed quantity and invited to support as long a list of features as
is at all possible within it;
Add-on A completely fresh piece of software is entitled to define its own file
formats and conventions and can generally be designed and build without
too much hindrance. But next year the extension to that original package
is needed, or the new program is one that has to work gracefully with data
from other people’s programs. When building an add-on it is painfully often
the case that the existing software base is not very well documented, and that
the attempted new use of it reveals previously unknown bugs or limitations
in the core system. Thus the effort that will need to be put into the second
package may be much greater than would have been predicted based on
experience from the first;
Embedded If the computer you are going to program is one in an electric eggtimer (or maybe a toy racing car, or an X-ray scanner) then testing may
involve be a quite different experience from that you become used to when
2 Last
year these notes indicates 16 Mbytes for £50! I may have rounded prices up and down
somewhat but still. . .
178
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
debugging ordinary applications that run on standard work-stations. In particular it may become necessary to become something of an expert in the
hardware and electronics and also in the application area of the system
within which your code will be placed;
Tool-weak environment This is a follow-on from the “embedded” heading, in
that it is perhaps easiest the envisage an electric pop-up toaster where anything that slowed down or enlarged the code being run would perturb system
timing enough to burn the toast, and where the target hardware is not automatically equipped with printers and flashing lights that can be used to help
sense what is going on inside its CPU. For some such cases it is possible
to buy or build real-time emulators or to wire in extra probes into a debuggable version of the hardware. There are other cases where either technology or budget mean that program development has to be done with a slow
turn-around on testing and with only very limited ability to discover what
happened when a bug surfaced. It is incredibly easy to simulate such a toolweak environment for yourself by just avoiding the effort associated with
becoming familiar with automated testing tools, debuggers and the like;
Novel One of the best and safest ways of knowing that a task is feasible is to
observe that somebody else did it before, and their version was at least more
or less satisfactory. The next best way is to observe that the new task is
really rather similar to one that was carried out successfully in the past. This
clearly leads to the obvious observation that if something is being attempted
and there are no precedents to rely on then it becomes much harder to predict
how well things will work out, and the chances of nasty surprises increases
substantially.
There are two sort of program not listed above which deserve special mention.
The first is the implementation of a known algorithm. This will usually end up as
a package or a subroutine rather than a complete free-standing program, and there
are plenty of algorithms that are complicated enough that programming them is
a severe challenge. However the availability of a clear target and well specified
direction will often make such programming tasks relatively tractable. It is however important to distinguish between programming up a complete and known
algorithm (easyish) from developing and then implementing a new one, and uncomfortably often things that we informally describe as algorithms are in fact just
strategies, and lots of difficult and inventive fine detail has to be filled into make
them realistic.
The second special sort of program is the little throw-away one, and the recognition that such programs can be lashed together really fast and without any fuss
5.2. ANALYSIS AND DESCRIPTION OF THE OBJECTIVE
179
is important, since it can allow one to automate other parts of the program development task through strategic use of such bits of jiffy code.
5.2 Analysis and description of the objective
Sometimes a programming task starts with you being presented with a complete,
precise and coherent explanation of exactly what has to be achieved. When this
is couched in language so precise that there is not possible doubt about what is
required you might like to ask why you are being asked to do anything, since
almost all you need to do is to transcribe the specification into the particular syntax
of the (specified) programming language. Several of the Part IA tickable problems
come fairly close to this pattern, and there the reason you are asked to do them
is exactly so you get practical experience with the syntax of the given language
and the practical details of presenting programs to the computer. But that hardly
counts as serious programming!
Assuming that we are not in one of these artificial cases, it is necessary to
think about what one should expect to find in a specification and what does not
belong there. It is useful to discuss the sorts of language used in specifications,
and to consider who will end up taking prime responsibility for everything being
correct.
A place to start is with the observation that a specification should describe
what is wanted, rather than how the desired effect is to be achieved. This ideal can
be followed up rather rapidly by the observation that it is often amazingly difficult
to know what is really wanted, and usually quite a lot of important aspects of
the full list of requirements will be left implicit or as items where you have to
apply your own judgement. This is where it is useful to think back to the previous
section and decide what style of project was liable to be intended and where the
main pressure points are liable to be.
5.2.1 Important Questions
I have already given a check-list that should help work out what general class of
problem you are facing. The next stage is to try to identify and concentrate on
areas of uncertainty in your understanding of what has to be done. Furthermore
initial effort ought to go into understanding aspects of the problem that are liable
to shape the whole project: there is no point in agonising over cosmetic details
until the big picture has become clear. Probably the best way of sorting this out is
to imagine that some magic wand has been waved and it has conjured up a body
of code and documentation that (if the fairy really was a good one!) probably
does everything you need. However as a hard-headed and slightly cynical person
180
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
you need to check it first. Deciding what you are going to look for to see if the
submitted work actually satisfied the project’s needs can let you make explicit a
lot of the previously slightly woolly expectations you might have. This viewpoint
moves you from the initial statement “The program must achieve X” a little closer
to “I must end up convinced that the program achieves X and here is the basis for
how that conviction might be carried”. Other things that might (or indeed might
not) reveal themselves at this stage are:
1. Is user documentation needed, and if so how detailed is it expected to be?
Is there any guess for how bulky the user manual will be?
2. How formal should documentation of the inner workings of the code be?
3. Was the implementation language to be used pre-specified, and if not what
aspects of the problem or environment are relevant to the choice?
4. Is the initial specification a water-tight one or does the implementer have to
make detailed design decisions along the way?
With regard to choice of programming language note that evidence from studies that have watched the behaviour of real programmers suggests that to a good
first approximation it is possible to deliver the same number of lines of working
documented code per week almost whatever language it is written in. A very
striking consequence of this is that languages that are naturally concise and which
provide built-in support for more of the high-level things you want to do can give
major boosts to productivity.
The object of all this thought is to lead to a proper specification of the task.
Depending on circumstances this may take one of a number of possible forms:
5.2.2 Informal specifications
Documents written in English, however pedantically phrased and however voluminous, must be viewed as informal specifications. Those who have a lot of spare
time might try reading the original description of the language C[16] where Appendix A is called a reference manual and might be expected to form a useful basis
for fresh implementations of the language. Superficially it looks pretty good, but
it is only when you examine the careful (though still “informal” in the current
context) description in the official ANSI standard[22] that it becomes clear just
how much is left unsaid in the first document. Note that ANSI C is not the same
language as that defined by Kernighan and Ritchie, and so the two documents just
mentioned can not be compared quite directly, and also be aware that spotting
and making clear places where specifications written in English are not precise
5.2. ANALYSIS AND DESCRIPTION OF THE OBJECTIVE
181
is a great skill, and one that some people enjoy exercising more than others do!
The description in section 5.18 is another rather more manageable example of an
informal textual specification. When you get to it you might like to check to see
what it tells you what to do about tabs and back-spaces, which are clearly characters that have an effect on horizontal layout. What? It fails to mention them? Oh
dear!
5.2.3 Formal descriptions
One response to the fact that almost all informal specifications are riddled with
holes (not all of which will be important: for instance it might be taken as understood by all that messages that are printed so that they look like sentences should
be properly spelt and punctuated) has been to look for ways of using formal description languages. The ZED language (developed at Oxford3, and sometimes
written as just Z) is one such and has at times been taught in Software Engineering courses here. The group concerned with the development of the language ML
were keen to use formal mathematically-styled descriptive methods to define exactly what ML ought to do in all possible circumstances. Later on in the CST
there are whole lecture courses on Specification and Verification and so I am not
going to give any examples here, but will content myself by observing that a good
grounding in discrete mathematics is an absolute pre-requisite for anybody thinking of working this way.
5.2.4 Executable specifications
One group of formal specification enthusiasts went off and developed ever more
powerful mathematical notations to help them describe tasks. Another group observed that sometimes a careful description of what must be achieved looks a
bit like a program in a super-compact super-high-level programming language.
It may not look like a realistic program, in that it may omit lots of explanation about how objectives should be achieved and especially how they should be
achieved reasonably efficiently. This leads to the idea of an executable specification, through building an implementation of the specification language. This will
permitted to run amazingly slowly, and its users will be encouraged to go all out
for clarity and correctness. To give a small idea of what this might entail, consider
the job of specifying a procedure to sort some data. The initial informal specification might be that the output should be a re-ordering of the input such that
the values in the output be in non-descending order. An executable specification
might consist of three components. The first would create a list of all the different
3 http://www.comlab.ox.ac.uk/oucl/prg.html
182
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
permutations of the input. The second would be a procedure to inspect a list and
check to see if its elements were in non-descending order. The final part would
compose these to generate all permutations then scan through them one at a time
and return the first non-descending one found. This would not be a good practical
sorting algorithm, but could provide a basis for very transparent demonstrations
that the process shown did achieve the desired goal! It should be remembered that
an executable specification needs to be treated as such, and not as a model for how
the eventual implementation will work. A danger with the technique is that it is
quite easy for accidental or unimportant consequences of how the specification is
written to end up as part of the project requirements.
5.3 Ethical Considerations
Quite early on when considering a programming project you need to take explicit
stack of any moral or ethical issues that it raises. Earlier in the year you have
had more complete coverage of the problems of behaving professionally, so here
I will just give a quick check-list of some of the things that might give cause for
concern:
1. Infringement of other people’s intellectual property rights, be they patents,
copyright or trade secrets. Some companies will at least try to prevent others
from creating new programs that look too much like the original. When
licensed software is being used the implications of the license agreement
may become relevant;
2. Responsibility to your employer or institution. It may be that certain sorts
of work are contrary to local policy. For instance a company might not be
willing to permit its staff to politically motivated virtual reality simulations
using company resources, and this University has views about the commercial use of academic systems;
3. A computing professional has a responsibility to give honest advice to their
“customer” when asked about the reasonableness or feasibility of a project,
and to avoid taking on work that they know they are not qualified to do;
4. It can be proper to take a considered stance against the development of
systems that are liable to have a seriously negative impact on society as a
whole. I have known some people who consider this reason to avoid any
involvement with military or defence-funded computing, while others will
object to technology that seems especially liable to make tracking, surveillance or eavesdropping easier. Those of you with lurid imaginations can no
5.4. HOW MUCH OF THE WORK HAS BEEN DONE ALREADY?
183
doubt envisage plenty more applications of computers that might be seen as
so undesirable that one should if necessary quit a job rather than work on
them.
5.4 How much of the work has been done already?
The points that I have covered so far probably do not feel as if they really help you
get started when faced with a hard-looking programming task, although I believe
that working hard to make sure you really understand the specification you are
faced with is in fact always a very valuable process. From now onwards I move
closer to the concrete and visible parts of the programming task. The first question
to ask here is “Do I actually have to do this or has it been done before?”
There are three notable cases where something has been done before but it is
still necessary to do it again. Student exercises are one of these, and undue reliance
on the efforts of your predecessors is gently discouraged. Sometimes a problem
has been solved before, but a solution needs to be re-created without reference to
the original version because the original is encumbered with awkward commercial4 restrictions or is not locally available. The final cause for re-implementation
is if the previous version of the program concerned was a failure and so much of a
mess that any attempt to rely on it would start the new project off in wrong-minded
directions.
Apart from these cases the best way to write any program at all is to adopt,
adapt and improve as much existing technology as you can! This can range from
making the very second program that you ever write a variation on that initial
“Hello World” example you were given through to exploiting existing large software libraries. The material that can be reclaimed may be as minor as a bunch of
initial comments saying who you (the author) are and including space to describe
what the program does. It might just be some stylised “import” statements needed
at the head of almost any program you write. If you need a tree structure in today’s
program do you have one written last week which gives you the data type definition and some of the basic operations on trees? Have you been provided with a
collection of nice neat sample programs (or do you have a book or CD ROM with
some) that can help? Many programming languages are packaged with a fairly
extensive collection of chunks of sample code.
Most programming languages come with standardised libraries that (almost always) mean there is no need for you to write your own sorting procedure or code to
convert floating point values into or out of textual form. In many important areas
4 Remember
that if the restriction is in the form of a patent then no amount of re-implementation
frees you from obligations to the patent-owner, and in other cases you may need to be able to give a
very clear demonstration that your new version really has been created completely independently.
184
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
there will be separate libraries that contain much much more extensive collections
of facilities. For instance numerical libraries (eg the one from NAG) are where
you should look for code to solve sets of simultaneous equations or to maximise
a messy function of several variables. When you need to implement a windowed
interface with pull-down menus and all that sort of stuff again look to existing
library code to cope with much of the low-level detail for you. Similarly for data
compression, arbitrary precision integer arithmetic, image manipulation. . .
Observing that there is a lot of existing support around does not make the
problem of program construction go away: knowing what existing code is available is not always easy, and understanding both how to use it and what constraints
must be accepted if it is used can be quite a challenge. For instance with the NAG
(numerical) library it may take beginners quite a while before they discover that
E04ACF (say, and not one of the other half-dozen closely related routines) is the
name of the function they need to call and before they understand exactly what
arguments to pass to it.
As well as existing pre-written code (either in source or library form) that
can help along with a new project there are also packages that write significant
bodies of code for you, basing what they do one on either a compact descriptive
input file or interaction with the user through some clever interface. The wellestablished examples of this are the tools yacc and lex that provide a convenient
and reliable way of creating parsers. Current users of Microsoft’s Visual C++
system will be aware of the so-called “Wizards” that it provides that help create
code to implement the user interface you want, and there are other commercial
program generators in this and a variety of business application areas. To use one
of these you first have to know of its availability, and then learn how to drive it:
both of these may involve an investment of time, but with luck that will be re-paid
with generous interest even on your first real use. In some cases the correct use
of a program generation tool is to accept its output uncritically, while on other
occasions the proper view is to collect what it creates, study it and eventually
adjust the generated code until you can take direct responsibility for subsequent
support. Before deciding which to do you need to come to a judgement about the
stability and reliability of the program generator and how often you will need to
adjust your code by feeding fresh input in to the very start.
Another way in which existing code can be exploited is when new code is
written so that it converts whatever input it accepts into the input format for some
existing package, one that solves a sufficiently related problem that this makes
some sense. For instance it is quite common to make an early implementation
of a new programming language work by translating the new language into some
existing one and then feeding the translated version into an existing compiler. For
early versions of ML the existing language was Lisp, while for Modula 3 some
compilers work by converting the Modula 3 source into C. Doing this may result
5.5. WHAT SKILLS AND KNOWLEDGE ARE AVAILABLE?
185
in a complete compiler that is slower and bulkier than might otherwise be the case,
but it can greatly reduce the effort in building it.
5.5 What skills and knowledge are available?
Figure 5.4: Nancy Silverton’s bakery is in La Brea near the tar pits, and her book
(Bread from the La Brea Bakery) is unutterably wonderful. I like her chocolatecherry loaf. This photo is if the racks in her shop. Not much about Java I agree
but baking good bread is at least as important to know about as computing.
A balance needs to be drawn between working through a programming project
using only the techniques and tools that you already know and pushing it forward
using valuable but unfamiliar new methods. Doing something new may slow you
down substantially, but an unwillingness to accept that toll may lead to a very
pedestrian style of code development using only a limited range of idioms. There
is a real possibility that short-term expediency can be in conflict with longer term
productivity. Examples where this may feel a strain include use of formal methods, new programming languages and program generation tools. The main point
to be got across here is that almost everything to do with computers changes every
five years or so, and so all in the field need to invest some of their effort in continual personal re-education so that their work does not look too much as if it has
been chipped out using stone axes. The good news is that although detailed technology changes the skills associated with working through a significant project
should grow with experience, and the amount of existing code that an old hand
will have to pillage may be quite large, and so there is a reasonable prospect for
a long term future for those with skills in software design and construction. Remember that all the books on Software Engineering tell us that the competence of
186
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
the people working on a project can make more difference to its success than any
other single factor.
It is useful to have a conscious policy of collecting knowledge about what
can be done and where to find the fine grubby details. For example the standard
textbook[9] contains detailed recipes for solving all sorts of basic tasks. Only
rarely will any one of these be the whole of a program you need to write, but quite
often a larger task will be able to exploit one or more of them. These and many of
the other topics covered in the CST are there because there is at least a chance that
they may occasionally be useful! It is much more important to know what can be
done than how to do it, because the how can always be looked up when you need
it.
5.6 Design of methods to achieve a goal
Perhaps the biggest single decision to be made when starting the detailed design
of a program is where to begin. The concrete suggestions that I include here are to
some extent caricatures; in reality few real projects will follow any of them totally
but all should be recognisable as strategies. The crucial issue is that it will not be
possible to design or write the whole of a program at once so it is necessary to
split the work into phases or chunks.
5.6.1 Top-Down Design
In Top Down Design work on a problem starts by writing a “program” that is just
one line long. Its text is:
{ solveMyProblem(); }
where of course the detailed punctuation may be selected to match the programming language being used. At this stage it is quite reasonable to be very informal
about syntax. A next step will be to find some way of partitioning the whole task
into components. Just how these components will be brought into existence is at
present left in the air, however if we split things up in too unreasonable a way we
will run into trouble later on. For many simple programs the second stage could
look rather like:
/* My name, today’s date, purpose of program */
import Standard-libraries;
{
/* declare variables here */
data = readInData();
results = calculate(data);
5.6. DESIGN OF METHODS TO ACHIEVE A GOAL
187
displayResults(results);
}
The ideal is that the whole development of the program should take place in
baby-sized steps like this. At almost every stage there will be a whole collection
of worrying-looking procedures that remain undefined and not yet thought about,
such as Calculate above. It is critical not to worry too much about these,
because each time a refinement is made although the number of these unresolved
problems may multiply the expected difficulty of each will reduce. Well it had
better, since all the ones that you introduce should be necessary steps towards
the solution of your whole original task, and it makes sense to expect parts to be
simpler than the whole.
After a rather few steps in the top-down development process one should expect to have a fully syntactically correct main program that will not need any
alterations later as the low level details of the procedures that it calls get sorted
out. And each of the components that remain to be implemented should have
a clearly understood purpose (for choice that should be written down) and each
such component should be clearly separated from all the others. That is not to
say that the component procedures might not call each other or rely on what they
each can do, but the internal details of any one component should not matter to
any other. This last point helps focus attention on interfaces. In my tiny example
above the serious interfaces are represented by the variables data and results
which pass information from one part of the design to the next. Working out exactly what must be captured in these interfaces would be generally need to be done
fairly early on. After enough stages of elaboration the bits left over from top-down
design are liable to end up small enough that you just code them up without need
to worry: anything that is trivial you code up, anything that still looks murky you
just apply one more expansion step to. With luck eventually the process ends.
There are two significant worries about top-down design. These are “How
do I know how to split the main task up?” and “But I can’t test me code until
everything is finished!”. Both of these are proper concerns.
Splitting a big problem up involves finding a strategy for solving it. Even
though this can be quite hard, it is almost always easier to invent a high-level idea
for how to solve a problem than it is to work through all the details, and this is
what top-down programming is all about. In many cases sketching on a piece of
paper what you would do if you had to solve the problem by hand (rather than
by computer) can help. Quite often the partition of a problem you make may end
up leading your design into some uncomfortable dead end. In that case you need
to look back and see which steps in your problem refinement represented places
where you had real choice and which ones were pretty much inevitable. It is then
necessary to go back to one of the stages where a choice was possible and to rethink things in the light of your new understanding. To make this process sensible
188
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
you should refuse to give up fleshing out one particular version of a top-down
design until you are in a position to give a really clear explanation of why the
route you have taken represents failure, because without this understanding you
will not know how far back you need to go in the re-planning. As an example
of what might go wrong, the code I sketched earlier here would end up being
wrongly structured if user interaction was needed, and that interaction might be
based on evaluation of partial results. To make that sort of interface possible it
might be necessary to re-work the design as (say)
/* My name, today’s date, purpose of program */
import Standard-libraries;
{
/* declare variables here */
/* set empty data and results */
while not finished do
{
extra = readInMoreData();
if (endOfUserInput(extra)) finished = true;
else
{
data = combine(data, extra);
results = upDateResults(results, data);
displaySomething(results);
}
}
displayFinalResults(results);
}
which is clearly getting messier! And furthermore my earlier and shorter version
looked generally valid for lots of tasks, while this one would need careful extra
review depending on the exact for of user interaction required.
There is a huge amount to be said in favour of being able to test a program as
it is built. Anybody who waits right to the end will have a dreadful mix of errors
at all possible levels of abstraction to try to disentagle. At first sight it seems that
top-down design precludes any early testing. This pessimism is not well founded.
The main way out is to write stubs of code that fill in for all the parts of your
program that have not yet been written. A stub is a short and simple piece of
code that takes the place of something that will later on be much more messy. It
does whatever is necessary to simulate some minimal behaviour that will make
it possible to test the code around it. Sometimes a stub will just print a warning
message and stop when it gets called! On other occasions one might make a stub
print out its parameters and wait for human intervention: it then reads something
back in, packages it up a bit and returns it as a result. The human assistant actually
5.6. DESIGN OF METHODS TO ACHIEVE A GOAL
189
did all the clever work.
There are two other attitudes to take to top-down design. One of these is to
limit it to design rather than implementation. Just use it to define a skeletal shape
for your code, and then make the coding and testing a separate activity. Obviously
this only makes sense when you have enough confidence that you can be sure that
the chunks left to be coded will in fact work out well. The final view is to think
of top-down design as an ideal to be retrofitted to any project once it is complete.
Even if the real work on a project went in fits and starts with lots of false trails
and confusion, there is a very real chance that it can be rationalised afterwards and
explained top-down. If that is done then it is almost certain that a clear framework
has been built for anybody who needs to make future changes to the program.
5.6.2 Bottom-Up Implementation
Perhaps you are uncertain about exactly what your program is going to do or how
it will solve its central problems. Perhaps you want to make sure that every line
of code you ever write is documented, tested and validated to death before you
move on from it and certainly before you start relying on it. Well these concerns
lead you towards a bottom-up development strategy. The idea here is to identify
a collection of smallish bits of functionality that will (almost) certainly be needed
as part of your complete program, and to start by implementing these. This avoids
having to thing about the hard stuff for a while. For instance a compiler-writer
might start by writing code to read in lines of program and discard comments,
or to build up a list of all the variable names seen. Somebody starting to write a
word processor might begin with pattern-matching code ready for use in searchand-replace operations. In almost all large projects there are going to be quite a
few fundamental units of code that are obviously going to be useful regardless of
the high level structure you end up with.
The worry with bottom-up construction is that it does not correspond to having
any overall vision of the final result. That makes it all to easy to end up with a
collection of ill-co-ordinated components that do not quite fit together and that do
not really combine to solve the original problem. At the very least I would suggest
a serious bout of top-down design effort be done before any bottom-up work to
try to put an overall framework into place. There is also a clear prospect that some
of the units created during bottom-up work may end up not being necessary after
all so the time spend on them was wasted.
An alternative way of thinking about bottom-up programming can soften the
impact of these worries. It starts by viewing a programming language not just as a
collection of fragments of syntax, but as a range of ways of structuring data and of
performing operations upon it. The fact that some of these operations happen to be
hard-wired into the language (as integer arithmetic usually is) while others exist as
190
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
collections of subroutines (floating point arithmetic on 3000-digit numbers would
normally be done that way) is of secondary importance. Considered this way each
time you define a new data type or write a fresh procedure you have extended and
customised your programming language by giving it support for something new.
Bottom-up programming can then be seen as gradually building layer upon layer
of extra support into your language until it is rich in the operations most important
in your problem area. Eventually one then hopes that the task that at first had
seemed daunting becomes just half a dozen lines in the extended language. If
some of the procedures built along the way do not happen to be used this time,
they may well come in handy the next time you have to write a program in the
same application area, so the work they consumed has not really been wasted
after all. The language Lisp is notable for having sustained a culture based on this
idea of language extension.
5.6.3 Data Centred Programming
Both top-down and bottom-up programming tend to focus on what your program
looks like and the way in which it is structured into procedures. An alternative is
to concentrate not on the actions performed by the code but on the way in which
data is represented and the history of transformations that any bit of data will be
subject to. These days this idea is often considered almost synonymous with an
Object Oriented approach where the overall design of the class structure for a program is the most fundamental feature that it will have. Earlier (and pre-dating the
widespread use of Object Oriented languages) convincing arguments for design
based on the entities that a program must manipulate or model come from Jackson
Structured Programming and Design[8]. More recently SSADM[3] has probably
become one of the more widespread design and specification methodologies for
commercial projects.
5.6.4 Iterative Refinement
My final strategy for organising the design of a complete program does not even
expect to complete the job in one session. It starts by asking how the initial problem can be restricted or simplified to make it easier to address. And perhaps it
will spot how the most globally critical design decisions for the whole program
could me made in two or three different ways, with it hard to tell in advance which
would work out best in the end. The idea is then to start with code for a scruffy
mock-up of a watered down version of the desired program using just one of these
sets of design decisions. The time and effort needed to write a program grows
much faster then linearly with the size of the program: the natural (but less obvious) consequence of this is that writing a small program can be much quicker and
5.7. HOW DO WE KNOW IT WILL WORK?
191
easier than completing the full version. It may in some cases make sense even to
write several competing first sketches of the code. When the first sketch version
is working it is possible to step back and evaluate it, to see if its overall shape is
sound. When it has been adjusted until it is structurally correct, effort can go into
adding in missing features and generally upgrading it until it eventually gets transformed into the beautiful butterfly that was really wanted. Of all the methods that
I have described this is the one that comes closest to allowing for “experimental”
programming. The discipline to adhere to is that experiments are worthy of that
tag if the results from them can be evaluated and if something can thus be learned
from them.
5.6.5 Which of the above is best?
The “best” technique for getting a program written will depend on its size as well
as its nature. I think that puritanical adherence to any of the above would be unreasonable, and I also believe that inspiration and experience (and good taste) have
important roles to play. However if pushed into an opinion I will vote for presenting a design or a program (whether already finished or still under construction)
as if it were prepared top-down, with an emphasis on the early design of what
information must be represented and where it must pass from one part of the code
to another.
5.7 How do we know it will work?
Nobody should ever write a program unless they have good reason to believe that
it ought to work. It is of course proper to recognise that it will not work, because
typographic errors and all sorts of oversights will ensure that. But the code should
have been written so that in slightly idealised world where these accidental imperfections do not exist it would work perfectly. Blind and enthusiastic hope is not
sufficient to make programs behave well, and so any proper design needs to have
lurking behind it the seeds of a correctness proof. In easy-going times this can remain untended as little comments that can just remind you of your thinking. When
a program starts to get troublesome it can be worth growing these comments into
short essays that explain what identities are being preserved intact across regions
of code, why your loops are guaranteed to terminate and what assumptions about
data are important, and why. In yet more demanding circumstances it can become
necessary to conduct formal validation procedures for code.
The easiest advice to give here is that before you write even half a dozen
lines of code you should write a short paragraph of comment that explains what
the code is intended to achieve and why your method will work. The comment
192
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
should usually not explain how it works (the code itself is all about “how”), but
why. To try to show that I (at least sometimes!) follow this advice here is a short
extract from one of my own5 programs. . .
/*
* Here is a short essay on the interaction between flags
* and properties. It is written because the issue appears
* to be delicate, especially in the face of a scheme that
* I use to speed things up.
* (a) If you use FLAG, REMFLAG and FLAGP with some
indicator then that indicator is known as a flag.
*
* (b) If you use PUT, REMPROP and GET with an indicator
then what you have is a property.
*
(c)
Providing the names of flags and properties are
*
disjoint no difficulty whatever should arise.
*
(d)
If you use PLIST to gain direct access to property
*
lists then flags are visible as pairs (tag . t) and
*
properties as (tag . value).
*
(e)
Using RPLACx operations on the result of PLIST may
*
cause system damage. It is considered illegal.
*
Also changes made that way may not be matched in
*
any accelerating caches that I keep.
*
* (f) After (FLAG ’(id) ’tag) [when id did not previously
have any flags or properties] a call (GET ’id ’tag)
*
will return t.
*
(g)
After (PUT ’id ’tag ’thing) a call (FLAGP ’id
*
’tag)
will return t whatever the value of "thing".
*
A call (GET ’id ’tag) will return the saved value
*
(which might be nil). Thus FLAGP can be thought of
*
as a function that tests if a given property is
*
attached to a symbol.
*
(h)
As a consequence of (g) REMPROP and REMFLAG are
*
really the same operation.
*
/
*
Lisp_Object get(Lisp_Object a, Lisp_Object b)
{
Lisp_Object pl, prev, w, nil = C_nil;
int n;
/*
5 This
happens to be written in C rather than Java, but since most of it is comment maybe that
does not matter too much.
5.7. HOW DO WE KNOW IT WILL WORK?
193
* In CSL mode plists are structured like association
* lists, and NOT as lists with alternate tags and values.
* There is also a bitmap that can provide a fast test for
* the presence of a property...
*/
if (!symbolp(a))
{
#ifdef RECORD_GET
record_get(b, NO);
errexit();
#endif
return onevalue(nil);
}
... etc etc
The exact details of what I am trying to do are not important here, but the evidence
of mind-clearing so that there is a chance to get the code correct first time is. Note
how little the comment before the procedure has to say about low-level implementation details, but how much about specifications, assumptions and limitations.
I would note here that typing a program in is generally one of the least timeconsuming parts of the whole programming process, and these days disc storage is
pretty cheap, and thus various reasons which in earlier days may have discouraged
layout and explanation in code no longer apply.
Before trying code and as a further check that it ought to work it can be useful
to “walk through” the code. In other words to pretend to be a computer executing
it and see if you follow the paths and achieve the results that you were supposed
to. While doing this it can be valuable to think about which paths through the
code are common and which are not, since when you get to testing it may be that
the uncommon paths do not get exercised very much unless you take special steps
to cause them to be activated.
The “correctness” that you will be looking for can be at several different levels. A partially correct program is one that can never give an incorrect answer.
This sounds pretty good until you recognise that there is a chance that it may just
get stuck in a loop and thereby never give any answer at all! It is amazingly often
much easier to justify that a program is partially correct than to go the whole hog
and show it is correct, ie that not only is it partially correct but that it will always
terminate. Beyond even the requirements of correctness will be performance demands: in some cases a program will need not only to deliver the right answers
but to meet some sort of resource budget. Especially if the performance target is
specified as being for performance that is good “on the average” it can be dreadfully hard to prove, and usually the only proper way to start is by designing and
justifying algorithms way before any mention of actual programming arises.
194
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
A final thing to check for is the possibility that your code can be derailed by
unhelpful erroneous input. For instance significant security holes in operating
systems have in the past been consequences of trusted modules of code being too
trusting of their input, and them getting caught out by (eg) input lines so long that
internal buffers overflowed thereby corrupting adjacent data.
The proper mind-set to settle in to while designing and starting to implement
code is pretty paranoid: you want the code to deliver either a correct result or a
comprehensible diagnostic whenever anything imaginable goes wrong in either
the data presented to it or its own internal workings. This last statement leads
to a concrete suggestion: make sure that the code can test itself for sanity and
correctness every so often and insert code that does just that. The assertions that
you insert will form part of your argument for why the program is supposed to
work, and can help you (later on) debug when it does not.
5.8 While you are writing the program
Please remember to get up and walk around, to stretch, drink plenty of water,
sit up straight and all the other things mentioned at the Learning Day as relevant
occupational health issues. My experience is that it is quite hard to do effective
programming in 5 minute snippets, but that after a few hours constant work productivity decreases. A pernicious fact is that you may not notice this decrease at
the time, in that the main way in which a programmer can become unproductive
is by putting more bugs into a program. It is possible to keep churning out lines of
code all through the night, but there is a real chance that the time you will spend
afterwards trying to mend the last few of them will mean that the long session did
not really justify itself.
In contrast to programming where long sessions can do real damage (because
of the bugs that can be added by a tired programmer) I have sometimes found that
long sessions have been the only way I can isolate bugs. Provided I can discipline
myself not to try to correct anything but the very simplest bug while I am tired
a long stretch can let me chase bugs in a painstakingly logical way, and this is
sometimes necessary when intuitive bug-spotting fails.
Thus my general advice about the concrete programming task would be to
schedule your time so you can work in bursts of around an hour per session, and
that you should plan your work so that as much as possible of everything you
do can be tested fairly enthusiastically while it is fresh in your mind. A natural
corollary of this advice is that projects should always be started in plenty of time,
and pushed forward consistently so that no last-minute panic can arise and force
sub-optimal work habits.
5.9. DOCUMENTING A PROGRAM OR PROJECT
195
5.9 Documenting a program or project
Student assessed exercises are expected to be handed in complete with a brief
report describing what has been done. Larger undergraduate projects culminate in
the submission of a dissertation, as do PhD studies. All commercial programming
activities are liable to need two distinct layers of documentation: one for the user
and one for the people who will support and modify the product in the future.
All these facts serve to remind us that documentation is an intrinsic part of any
program.
Two overall rules can guide the writing of good documentation. The first is
to consider the intended audience, and think about what they need to know and
how your document can be structured to help them find it. The second is to keep a
degree of consistency and order to everything: documents with a coherent overall
structure are both easier to update and to browse than sets of idiosyncratic jottings.
To help with the first of these, here are some potential styles of write-up that
might be needed:
1. Comments within the code to remind yourself or somebody who is already
familiar with the program exactly what is going on at each point in it;
2. An overview of the internal structure and organisation of the whole program
so that somebody who does not already know it can start to find their way
around;
3. Documentation intended to show how reliable a program is, concentrating
on discussions of ways in which the code has been built to be resilient in the
face of unusual combinations of circumstance;
4. A technical presentation of a program in a form suitable for publication in a
journal or at a conference, where the audience will consist of people expert
in the general field but not aware of exactly what your contribution is;
5. An introductory user manual, intended to make the program usable even by
the very very nervous;
6. A user reference manual, documenting clearly and precisely all of the options and facilities that are available;
7. On-line help for browsing by the user while they are trying to use the program;
8. A description of the program suitable for presentation to the venture capitalists who are considering investing in the next stage of its development.
196
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
It seems inevitable that the above list is not exhaustive, but my guess is that
most programs could be presented in any one of the given ways, and the resulting
document would be quite different in each case. It is not that one or the other of
these styles is inherently better or more important than another, more that if you
write the wrong version you will either not serve your reader well or you will find
that you have had to put much more effort into the documentation than was really
justified.
A special problem about documentation is that of when it should be written.
For small projects at least it will almost always be produced only after the program
has been (at least nearly) finished. This can be rationalised by claiming “how can
I possibly document it before it exists?”
I will argue here for two ideals. The first is that documentation ought to follow on from design and specification work, but precede detailed programming.
The second is that the text of the documentation should live closely linked to the
developing source code. The reasoning behind the first of these is that writing
the text can really help to ensure that the specification of the code has been fully
thought through, and once it is done it provides an invaluable stable reference to
keep the detailed programming on track. The second point recognises some sort
of realism, and that all sorts of details of just what a program does will not be
resolved until quite late in the implementation process. For instance the exact
wording of messages that are printed will often not be decided until then, and it
will certainly be hard to prepare sample transcripts from the use of the program
ahead of its completion6. Thus when the documentation has been written early it
will need completing when some of these final details get settled and correcting
when the code is corrected or extended. The most plausible way of making it
feasible to keep code and description in step is to keep them together. The concept of Literate Programming[17] pursues this goal. A program is represented as
a composite file that can be processed in (at least) two different ways. One way
“compiles” it to create typeset-quality human readable documentation, while the
other leaves just statements in some quite ordinary programming language ready
to be fed into a compiler. This goes beyond just having copious comments in the
code in two ways. Firstly it expects that the generated documentation should be
able to exploit the full range of modern typography and that it can include pictures or diagrams where relevant. It is supposed to end up as cleanly presented
and readable as any fully free-standing document could ever be. Secondly Literate
Programming recognises that the ordering and layout of the program that has to
be compiled may not be the same as that in the ideal manual, and so the disentangling tool needs to be able to rearrange bits of text in a fairly flexible way so that
description can simultaneously be thought of as close to the code it relates to and
6 Even
though these samples can be planned and sketched early.
5.10. HOW DO WE KNOW IT DOES WORK?
197
to the section in the document where it belongs. This idea was initially developed
as part of the project to implement the TEX typesetting program that is being used
to prepare these lecture notes.
5.10 How do we know it does work?
Figure 5.5: Many people think that their work is over well before it actually is.
A conceptual difficulty that many people suffer from is a confusion between
whether a program should work and whether it does. A program should work if
it has been designed so that there are clear and easily explained reasons why it
can achieve what it should. Sometimes the term “easily explained” may conceal
the mathematical proof of the correctness of an algorithm, but at least in theory it
198
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
would be possible to talk anybody through the justification. As to programs that
actually do work, well the reality seems to be that the only ones of these that you
will ever see will be no more than around 100 lines long: empirically any program
much longer than that will remain flawed even after extensive checking. Proper
Oriental rugs will always have been woven with a deliberate mistake in them,
in recognition of the fact that only Allah is perfect. Experience has shown very
clearly indeed that in the case of writing programs we all have enough failings
that there is no great need to insert extra errors — there will be plenty inserted
however hard we try to avoid them. Thus (at least at the present state of the art)
there is no such thing as a (non-trivial) program that works.
If, however, a program should work (in the above sense) then the residual
errors in it will be ones that can be corrected without disturbing the concepts
behind it or its overall structure. I would like to think of such problems as “little
bugs”. The fact that they are little does not mean that they might not be important,
in that missing commas or references to the wrong variable can cause aeroplanes
to crash just as convincingly as can errors at a more conceptual level. But the
big effort must have been to get to a first testable version of your code with only
little bugs left in it. What is then needed is a testing strategy to help locate as
many of these as possible. Note of course that testing can only ever generate
evidence for the presence of a bug: in general it can not prove absence. But
careful and systematic testing is something we still need whenever there has been
human involvement in the program construction process7 .
The following thoughts may help in planning a test regime:
1. Even obvious errors in output can be hard to notice. Perhaps human society
has been built up around a culture of interpreting slightly ambiguous input
in the “sensible” way, and certainly we are all very used to seeing what we
expect to see even when presented with something rather different. By the
time you see this document I will have put some effort into checking its
spelling, punctuation, grammar and general coherence, and I hope that you
will not notice or be upset by the residual mistakes. But anybody who has
tried serious proof-reading will be aware that blatant mistakes can emerge
even when a document has been checked carefully several times;
2. If you are checking your own code and especially if you know you can
stop work once it is finished then you have a clear incentive not to notice
mistakes. Even if a mistake you find is not going to cause you to have to
spend time fixing it it does represent you having found yet another instance
of your own lack of attention, and so it may not be good for your ego;
7 Some
see this observation as a foundation for hope for the future
5.10. HOW DO WE KNOW IT DOES WORK?
199
3. It is very desirable to make a clear distinction between the job of testing a
program to identify the presence of bugs and the separate activity of correcting things. It can be useful to take the time to try to spot as many mistakes
as you can before changing anything at all;
4. A program can contain many more bugs and oddities than your worst nightmares would lead you to believe!
5. Testing strategies worked out as part of the initial design of a program are
liable to be better than ones invented only once code has been completed;
6. It can be useful to organise explicit test cases for extreme conditions that
your program may face (eg sorting data where all the numbers to be sorted
have the same value), and to collect test cases that cause each path through
your code to be exercised. It is easy to have quite a large barrage of test
cases but still have some major body of code unvisited.
7. Regressions tests are a good thing. These are test cases that grow up during
project development, and at each stage after any change is made all of them
are re-run, and the output the produce is checked. When any error is detected a new item in the regression suite is prepared so that there can remain
a definite verification that the error does not re-appear at some future stage.
Automating the application of regression tests is a very good thing, since
otherwise laziness can too easily cause one to skip running them;
8. When you find one bug you may find that its nature gives you ideas for other
funny cases to check. You should try to record your thoughts so that you do
not forget this insight;
9. Writing extra programs to help you test your main body of code is often a
good investment in time. On especially interesting scheme is to generate
pseudo-random test cases. I have done that while testing a polynomial factorising program and suffered randomly-generated tests of a C compiler I
was involved with, and in each case the relentless random coverage of cases
turned out to represent quite severe stress;
10. You do not know how many bugs your code has in it, so do not know when
to stop looking. One theoretical way to attack this worry would be to get
some fresh known bugs injected into your code before testing, and then see
what proportion of the bugs found were the seeded-in ones and which had
been original. That may allow you to predict the total bug level remaining.
Having detected some bugs there are several possible things to do. One is to sit
tight and hope that nobody else notices! Another is to document the deficiencies
200
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
at the end of your manual. The last is to try to correct some of them. The first
two of these routes are more reasonable than might at first seem proper given that
correcting bugs so very often introduces new ones.
In extreme cases it may be that the level of correctness that can be achieved
by bug-hunting will be inadequate. Sometimes it may then be possible to attempt
a formal proof of the correctness of your code. In all realistic circumstances this
will involve using a large and complicated proof assistant program to help with all
the very laborious details involved. Current belief is that it will be very unusual
for bugs in the implementation of this tool to allow you to end up with a program
that purports to be proved but which in fact still contains mistakes!
5.11 Is it efficient?
I have made this a separate section from the one on detecting the presence of
errors because performance effects are only rarely the result of simple oversights.
Let me start by stressing the distinction between a program that is expensive to
run (eg the one that computes π to 20,000,000,000 decimal places) and ones that
are inefficient (eg one that takes over half a second to compute π correct to four
places). The point being made is that unless you have a realistic idea of how long a
task ought to take it is hard to know if your program is taking a reasonable amount
of time. And similarly for memory requirements, disc I/O or any other important
resource. Thus as always we are thrown back to design and specification time
predictions as our only guideline, and sometimes even these will be based on little
more than crude intuition.
If a program runs fast enough for reasonable purposes then there may be no
benefit in making it more efficient however much scope for improvement there is.
In such cases avoid temptation. It is also almost always by far best to concentrate
on getting code correct first and only worry about performance afterwards, taking
the view that a wrong result computed faster is still wrong, and correct results may
be worth waiting for.
When collecting test cases for performance measurements it may be useful
to think about whether speed is needed in every single case or just in most cases
when the program is run. It can also be helpful to look at how costs are expected to
(and do) grow as larger and larger test cases are attempted. For most programming
tasks it will be possible to make a trade between the amount of time a program
takes to run and the amount of memory it uses. Frequently this shows up in a
decision as to whether some value should be stored away in case it is needed
later or whether any later user should re-calculate it. Recognising this potential
trade-off is part of performance engineering.
For probably the majority of expensive tasks there will be one single part of the
5.12. IDENTIFYING ERRORS
201
entire program that is responsible for by far the largest amount of time spent. One
would have expected that it would always be easy to predict ahead of time where
that would be, but it is not! For instance when an early TITAN Fortran compiler
was measured in an attempt to discover how it could be speeded up it was found
that over half of its entire time was spent in a very short loop of instructions that
were to do with discarding trailing blanks from the end of input lines. Once the
programmers knew that it was easy to do something about it, but one suspects
they were expecting to find a hot-spot in some more arcane part of the code. It
is thus useful to see if the languages and system you use provide instrumentation
that makes it easy to collect information to reveal which parts of your code are
most critical. If there are no system tools to help you you may be able to add in
time-recording statements to your code so it can collect its own break-down to
show what is going on. Cunning optimisation of bits of code that hardly ever get
used is probably a waste of effort.
Usually the best ways to gain speed involve re-thinking data structures to provide cheap and direct support for the most common operations. This can sometimes mean replacing a very simple structure by one that has huge amounts of algorithmic complexity (there are examples of such cases in the Part IB Complexity
course and the Part II one on Advanced Algorithms). In almost all circumstances
a structural improvement that gives a better big-O growth rate for some critical
cost is what you should seek.
In a few cases the remaining constant factor improvement in speed may still
be vital. In such cases it may be necessary to re-write fragments of your code in
less portable ways (including the possibility of use of machine code) or do other
things that tend to risk the reliability of your package. The total effort needed to
complete a program can increase dramatically as the last few percent in absolute
performance gets squeezed out.
5.12 Identifying errors
Section 5.7 was concerned with spotting the presence of errors. Here I want to talk
about working out which part of your code was responsible for them. The sections
are kept separate to help you to recognise this, and hence to allow you to separate
noticing incorrect behaviour from spotting mistakes in your code. Of course if,
while browsing code, you find a mistake you can work on from it to see if it can
ever cause the program to yield wrong results, and this study of code is one valid
error-hunting activity. But even in quite proper programs it is possible to have
errors that never cause the program to misbehave in any way that can be noticed.
For instance the mistake might just have a small effect on the performance of
some not too important subroutine, or it may be an illogicality that could only
202
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
be triggered into causing real trouble by cases that some earlier line of code had
filtered out.
You should also recognise that some visible bugs are not so much due to any
single clear-cut error in a program but to an interaction between several parts of
your code each of which is individually reasonable but which in combination fail.
Most truly serious disasters caused by software failure arise because of complicated interactions between multiple “improbable” circumstances.
The first thing to try to locate the cause of an error is to start from the original
test case that revealed it and to try to refine that down to give a minimal clear-cut
demonstration of the bad behaviour. If this ends up small enough it may then be
easy to trace through and work out what happened.
Pure thought and contemplation of your source code is then needed. Decide
what Sherlock Holmes would have made of it! Run your compilers in whatever
mode causes them to give as many warning messages as they are capable of, and
see if any of those give valuable clues. Check out the assert facility and place
copious assertions in your program that verify that all the high level expectations
you have are satisfied.
If this fails the next thought is to arrange to get a view on the execution of
your code as it makes its mistake. Even when clever language-specific debuggers
are available it will often be either necessary or easiest to do this by extra print
statements into your code so it can display a trace of its actions. There is a great
delicacy here. The trace needs to be detailed enough to allow you to spot the first
line in it where trouble has arisen, but concise enough to be manageable. My
belief is that one should try to judge things so that the trace output from a failing
test run is about two pages long.
There are those who believe that programs will end up with the best reliability
if they start off written in as fragile way as possible. Code should always make as
precise a test as possible, and should frequently include extra cross checks which,
if failed, cause it to give up. The argument is that this way a larger number of
latent faults will emerge in early testing, and the embedded assertions can point
the programmer directly to the place where an expectation failed to be satisfied,
which is at least a place to start working backwards from in a hunt for the actual
bug.
5.12. IDENTIFYING ERRORS
Figure 5.6: Effective debugging calls for great skill.
203
204
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
With many sorts of bugs it can be possible to home in on the difficulty by some
sort of bisection search. Each test run should be designed to halve the range of
code within which the error has been isolated.
Some horrible problems seem to vanish as soon as you enable any debugging
features in your code or as soon as you insert extra print statements into it. These
can be amazingly frustrating! They may represent your use of an unsafe language
and code that writes beyond the limit of an array, or they could involve reliance on
the unpredictable value of an un-initialised variable. Sometimes such problems
turn out to be bugs in the compiler you are using, not in your own code. I believe
that I have encountered trouble of some sort (often fairly minor, but trouble nevertheless) with every C compiler I have ever used, and I have absolute confidence
that no other language has attained perfection in this regard. So sometimes trying
your code on a different computer or with a different compiler will either give you
a new diagnostic that provides the vital clue, or will behave differently thereby
giving scope for debugging-by-comparison.
Getting into a panic and trying random changes to your code has no proper
part to play either in locating or identifying bugs.
5.13 Corrections and other changes
With a number of bugs spotted and isolated the time comes to extirpate them. The
ideal should be that when a bug is removed it should be removed totally and it
should never ever be able to come back. Furthermore its friends and offspring
should be given the same treatment at the same time, and of course no new mistakes should be allowed to creep in while the changes are being made. This last
is often taken for granted, but when concentrating on one particular bug it is all
too easy to lose sight of the overall pattern of code and even introduce more new
bugs than were being fixed in the first case. Regression testing is at least one line
of defence that one should have against this, but just taking the correction slowly
and thinking through all its consequences what is mostly wanted. Small bugs (in
the sense discussed earlier) that are purely local in scope give fewest problems.
However sometimes testing reveals a chain of difficulties that must eventually be
recognised as a sign that the initial broad design of the program had been incorrect, and that the proper correction strategy does not involve fixing the problems
one at a time but calls for an almost fresh start on the whole project. I think that
would be the proper policy for the program in section 5.18, and that is part of why
the exercise there asks you to identify bugs but not to correct them.
Upgrading a program to add new features is at least as dangerous as correcting
bugs, but in general any program that lasts for more than a year or so will end up
with a whole raft of alterations having been made to it. These can very easily
5.14. PORTABILITY OF SOFTWARE
205
damage its structure and overall integrity, and the effect can be thought of as a
form of software rot that causes old code to decay. Of course software rot would
not arise if a program never needed correcting and never needed upgrading, but
in that case the program was almost certainly not being used and was fossilised
rather than rotting. Note that for elderly programs the person who makes corrections is never the original program author (even if they have the same name and
birthday, the passage of time has rendered them different). This greatly increases
the prospect of a would-be correction causing damage.
All but the most frivolous code should be kept under the control of some source
management tool (perhaps rcs) that can provide an audit trail so that changes can
be tracked. In some cases a discussion of a bug that has now been removed might
properly remain as a comment in the main source code, but much more often a
description of what was found to be wrong and what was changed to mend it
belongs in a separate project log. After all if the bug really has been removed who
has any interest in being reminded of the mistake that it represented?
Whenever a change is made to a program, be it a bug-fix or an upgrade, there
is a chance that some re-work will be needed in documentation, help files, sample
logs and of course the comments. Once again the idea of literate programming
comes to the fore in suggestion that all these can be kept together.
5.14 Portability of software
Most high level languages make enthusiastic claims that programs written in them
will be portable from one brand of computer to another, just as most make claims
that their compilers are “highly optimising”. Java makes especially strong claims
on this front, and its owners try rather hard to prevent anybody from diverging
from a rigid standard. However even in this case there are differences between
Java 1.0 and 1.1 (and no doubt 1.2) that may cause trouble to the unwary.
In reality achieving portability for even medium sized programs is not as easy
as all that. To give a gross example of a problem not addressed at all by programming language or standard library design, a Macintosh comes as standard with
a mouse with a single button, while most Unix X-windows systems have threebutton mice. In one sense the difference is a frivolity, but at another it invites a
quite substantial re-think of user interface design. At the user interface level a design that makes good use of a screen with 640 by 480 pixels and 16 or 256 colours
(as may be the best available on many slightly elderly computers) may look silly
on a system with very much higher resolution and more colours.
For most programming languages you will find that implementations provided
by different vendors do not quite match. Even with the most standardised languages hardly any compiler supplier will manage to hold back from providing
206
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
some private extra goodies that help distinguish them from their competitors. Such
extras will often be things that it is very tempting to make use of. Around 1997-8
a good example of such a feature is “Active-X” which Microsoft is promoting. To
use such a feature tends to lock you to one vendor or platform, while to ignore it
means that you can not benefit from the advantages that it brings. By now you will
know what my suggested response to conflicts like this will be. Yes, it is to make
your decisions explicitly and consciously rather than by default, to make them in
view of stated ideas about what the users of your code will need, and to include
all the arguments you use to support your decision in your design portfolio.
There are frequently clever but non-portable tricks that can lead to big performance gains in code but at cost in portability. Sometimes the proper response to
these is to have two versions of the program, one slow but very portable and the
other that takes full advantage of every trick available on some platform that is
especially important to you.
5.15 Team-work
Almost all of this course is about programming in the small, with a concentration
on the challenges facing a lone programmer. It is still useful to think for a while
how to handle the transition from this state into a large-team corporate mentality.
One of the big emotional challenges in joining a team relates to the extent to
which you end up “owning” the code you work on. It is very easy to get into a
state where you believe (perhaps realistically) that you are the only person who
can properly do anything to the code you write. It is also easy to become rather
defensive about your own work. A useful bit of jargon that refers to breaking out
of these thought patterns is ego-free programming. In this ideal you step back and
consider the whole project as the thing you are contributing to, not just the part
that you are visibly involved in implementing. It may also be useful to recognise
that code will end up with higher quality if understanding of it is shared between
several people, and that bugs can be viewed as things to be found and overcome
and never as personal flaws in the individual who happened to write that fragment
of code.
When trying to design code or find a difficult bug it can be very valuable to
explain your thoughts to somebody else. It may be that they need not say much
more than er and um, and maybe they hardly need to listen (but you probably need
to believe that they are). By agreeing that you will listen to their problems at a
later stage this may be a habit you can start right now with one or a group of your
contemporaries.
Reading other people’s code (with their permission, of course) and letting
them read yours can also help you settle on a style or idiom that works well for
5.16. LESSONS LEARNED
207
you. It can also help get across the merits of code that is well laid out and where
the comments are actually helpful to the reader.
If you get into a real group programming context, it may make sense to consider partitioning the work in terms of function, for instance system architect,
programmer, test case collector, documentation expert,. . . rather than trying to distribute the management effort and split the programming into lots of little modules, but before you do anything too rash read some more books on software engineering so that once again you can make decisions in an informed and considered
way.
5.16 Lessons learned
One of the oft-repeated observations about the demons of large-scale software
construction is that there is no silver bullet. In other words we can not expect to
find a single simple method that, as if by magic, washes away all our difficulties.
This situation also applies for tasks that are to be carried out by an individual
programmer or a very small team. No single method gives a key that makes it
possible to sit down and write perfect programs without effort. The closest I can
come to an idea for something that is generally valuable is experience – experience
on a wide range of programming projects in several different languages and with
various different styles of project. This can allow you to spot features of a new
task that have some commonalty with one seen before. This is, however, obviously
no quick fix. The suggestions I have been putting forward here are to try to make
your analysis of what you are trying to achieve as explicit in your mind as possible.
The various sections in these notes provide headings that may help you organise
your thoughts, and in general I have tried to cover topics in an order that might
make sense in real applications. Of course all the details and conclusions will be
specific to your problem, and nothing I can possibly say here can show you how
to track down your own very particular bug or confusion! I have to fall back on
generalities. Keep thinking rather than trying random changes to your code. Try
to work one step at a time. Accept that errors are a part of the human condition,
and however careful you are your code will end up with them.
But always remember the two main slogans:
Programming is easy
and
Programming is fun.
208
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
5.17 Final Words
Do I follow my own advice? Hmmm I might have known you would ask that!
Well most of what I have written about here is what I try to do, but I am not
especially formal about any of it. I only really go overboard about design and
making documentation precede implementation when starting some code that I
expect to give me special difficulty. I have never got into the swing of literate
programming, and suspect that I like the idea more than the reality. And I sometimes spend many more hours on a stretch at a keyboard than I maybe ought to. If
this course and these notes help you think about the process of programming and
allow you to make more conscious decisions about the style you will adopt then
I guess I should be content. And if there is one very short way I would like to
encapsulate the entire course, it would be the recommendation that you make all
the decisions and thoughts you have about programming as open and explicit as
possible.
Good luck!
5.18 Challenging exercises
Some of you may already consider yourselves to be seasoned programmers able
to cope with even quite large and complicated tasks. In which case I do not you to
feel this course is irrelevant, and so I provide here at the end of the notes some programming problems which I believe are hard enough to represent real challenges,
even though the code that eventually has to be written will not be especially long.
There is absolutely no expectation that anybody will actually complete any of
these tasks, or even find good starting points. However these examples may help
give you concrete cases to try out the analysis and design ideas I have discussed:
identifying the key difficulties and working out how to partition the problems into
manageable chunks. In some cases the hardest part of a proper plan would be the
design of a good enough testing strategy. The tasks described here are all both
reasonably compact and fairly precisely specified. I have fought most of these
myself and found that producing solutions that were neat and convincing as well
as correct involved thought as well as more coding skill. There are no prizes and
no ticks, marks or other bean-counter’s credit associated with attempting these
tasks, but I would be jolly interested to see what any of you can come up with,
provided it can be kept down to no more than around 4 sides of paper.
5.18. CHALLENGING EXERCISES
209
MULDIV
The requirement here is to produce a piece of code that accepts four integers and
computes (a ∗ b + c)/d and also the remainder from the division. It should be
assumed that the computer on which this code is to be run has 32-bit integers, and
that integer arithmetic including shift and bitwise mask operations are available,
but the difficulty in this exercise arises because a ∗ b will be up to 64-bits long
and so it can not be computed directly. “Solutions” that use (eg) the direct 64-bit
integer capabilities of a DEC Alpha workstation are not of interest!
It should be fairly simple to implement muldiv if efficiency where not an
issue. To be specific this would amount to writing parts of a package that did
double-length integer arithmetic. Here the additional expectation is that speed
does matter, and so the best solution here will be one that makes the most effective possible use of the 32-bit arithmetic that is available. Note also that code of
this sort can unpleasantly easily harbour bugs, for instance due to some integer
overflow of an intermediate result, that only show up in very rare circumstances,
and that the pressure to achieve the best possible performance pushes towards code
that comes very close to the limits of the underlying 32-bit arithmetic. Thought
will be needed when some or all of the input values are negative. The desired behaviour is one where the calculated quotient was rounded towards zero, whatever
its sign.
Overlapping Triangles
A point in the X–Y plane can be specified by giving its co-ordinates (x, y). A
triangle can then be defined by giving three points. Given two triangles a number
of possibilities arise: they may not overlap at all or they may meet in a point or
a line segment, or they may overlap so that the area where they overlap forms a
triangle, a quadrilateral, a pentagon or a hexagon. Write code that discovers which
of these cases arises, returning co-ordinates that describe the overlap (if any).
A point to note here is that any naive attempt to calculate the point where
two lines intersect can lead to attempts to divide by zero if the lines are parallel.
Near-parallel lines can lead to division by very small numbers, possibly leading
to subsequent numeric overflow. Such arithmetic oddities must not be allowed to
arise in the calculations performed.
Matrix transposition
One way of representing an m by n matrix in a computer is to have a single vector
of length mn and place the array element ai, j at offset mi + j in the vector. Another
would be to store the same element at offset i + n j. One of these representation
210
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
means that items in the same row of the matrix live close together, the other that
items in the same column are adjacent.
In some calculations it can make a significant difference to speed which of
these layouts is used. This is especially true for computers with virtual memory.
Sometimes one part of a calculation would call for one layout, and a later part
would prefer the other.
The task here is therefore to take integers m and n and a vector of length mn,
and rearrange the values stored in the vector so that if they start off in one of as
one representation of a matrix they end up as the other. Because the matrix should
be assumed to be quite large you are not allowed to use any significant amount of
temporary workspace (you can not just allocate a fresh vector of length mn and
copy the data into it in the new order — you may assume you may use extra space
of around m + n if that helps, but not dramatically more than that).
If the above explanation of the problem8 feels out of touch with today’s computer uses, note how the task relates to taking an m by n matrix representing a picture and shuffling the entries to get the effect of rotating the image by 90 degrees.
Just that in the image processing case you may be working with data arranged in
sub-word-sized bite-fields, say at 4 bits per pixel.
Sprouts
The following is a description of a game9 to be played by two players using a
piece of paper. The job of the project here is to read in a description of a position
in the game and make a list of all the moves available to the next player. This
would clearly be needed as part of any program that played the game against
human opposition, but the work needed here does not have to consider any issues
concerning the evaluation of positions or the identification of good moves.
The game starts with some number of marks made on a piece of paper, each
mark in the form of a capital ‘Y’. Observe that each junction has exactly three
little edges jutting from it. A move is made by a player identifying two free edges
and drawing a line between them. The line can snake around things that have been
drawn before as much as the player making the move likes, but it must not cross
any line that was drawn earlier. The player finishes the move by drawing a dot
somewhere along the new line and putting the stub of a new edge jutting out from
it in one of the two possible directions. Or put a different but equivalent way, the
player draws a new ‘Y’ and joins two of its legs up to existing stubs with lines that
do not cross any existing lines. The players make moves alternately and the first
player unable to make a further legal move will be the loser.
8 This
is an almost standard classical problem and if you dig far enough back in the literature
you will find explanations of a solution. If you thought to do that for yourself, well done!
9 Due to John Conway
5.18. CHALLENGING EXERCISES
211
A variation on the game has the initial state of the game just dots (not ‘Y’
shapes) and has each player draw a new dot on each edge they create, but still
demands that no more that three edges radiate from each dot. The difference
is that in one case a player can decide which side of a new line any future line
must emerge from. I would be equally happy whichever version of the game
was addressed by a program, provided the accompanying documentation makes it
clear which has been implemented!
The challenge here clearly largely revolves around finding a way to describe
the figures that get drawn. If you want to try sprouts out as a game between people
before automating it, I suggest you start with five or six starting points.
ML development environment
The task here is not to write a program, but just to sketch out the specification
of one. Note clearly that an implementation of the task asked about here would
be quite a lot of work and I do not want to provide any encouragement to you to
attempt all that!
In the Michaelmas Term you were introduced to the language ML, and invited
to prepare and test various pieces of test code using a version running under Microsoft Windows. You probably used the regular Windows “notepad” as a little
editor so you could change your code and then paste back corrected versions of
it into the ML window. Recall that once you have defined a function or value in
ML that definition remains fixed for ever, and so if it is incorrect you probably
need to re-type not only it but everything you entered after it. All in all the ML
environment you used was pretty crude (although I am proud of the greek letters
in the output it generates), and it would become intolerable for use in medium or
large-scale projects. Design a better environment, and append to your description of it a commentary about which aspects of it represent just a generically nice
programmer’s work-bench and which are motivated by the special properties of
ML.
An example from the literature
The following specification is given as a paragraph of reasonably readable English
text, and there is then an associated program written in Java. This quite small
chunk of code can give you experience of bug-hunting: please do not look up the
original article in CACM10 until you have spent some while working through the
code checking how it works and finding some of the mistakes. In previous years
when I have presented this material to our students they did almost as well as the
professional programmers used in the original IBM study, but they still found only
10 Communications of
the ACM, vol 21, no 9, 1978.
212
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
a pathetically small proportion of the total number of known bugs! I am aware that
the Java transcription of this program has changed its behaviour from the original
PL/I version and the intermediate C one. I do not believe that the changes are a
case of bugs evaporating in the face of Java, but some may be transcription errors
I have made. But since the object of this exercise is that you locate bugs, whatever
their source, this does not worry me much, and I present the example as it now
stands, oddities and all.
Figure 5.7: Picture courtesy Shamila Corless.
5.18. CHALLENGING EXERCISES
Formatting program for text input. Converted from
the original PL/I version which is in a paper by Glen
Myers, CACM vol 21 no 9, 1978
(a) This program compiles correctly: it is believed
not to contain either syntax errors or abuses of
the Java library.
(b) A specification is given below. You are to imagine
that the code appended was produced by somebody who
had been provided with the specification and asked
to produce an implementation of the utility as
described. But they are not a very good programmer!
(c) Your task is one of quality control - it is to
check that the code as given is in agreement with
the specification.
If any bugs or mis-features are discovered they
should be documented but it will be up to the
original programmer to correct them. If there are
bugs it is desirable that they all be found.
(d) For the purposes of this study, a bug or a
mis-feature is some bad aspect of the code that
could be visible to users of the binary version of
the code. Ugly or inefficient code is deemed not
to matter, but even small deviations from the
letter of the specification and the things sensibly
implicit in it do need detecting.
(e) Let me repeat point (a) again just to stress it the code here has had its syntax carefully checked
and uses the Java language and library in a legal
straightforward way, so searching for bugs by
checking fine details of the Java language
specification is not expected to be productive.
I have put in comments to gloss use of library
functions to help those who do not have them all
at their finger-tips. The code may be clumsy in
places but I do not mind that!
I have tried to keep layout of the code neat and
consistent. There are few comments "because the
original programmer who wrote the code delivered
it in that state".
213
214
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
/*********************************************************
* Specification
*
* =============
*
*
*
Given
an
input
text
consisting
of
words
separated
by
*
*
* blanks or new-line characters, the program formats it *
* into a line-by-line form such that (1) each output
*
line
has
a
maximum
of
30
characters,
(2)
a
word
in
*
*
* the input text is placed on a single output line, and *
* (3) each output line is filled with as many words as *
* possible.
*
*
*
The
input
text
is
a
stream
of
characters,
where
the
*
*
* characters are categorised as break or nonbreak
*
characters.
A
break
character
is
a
blank,
a
new-line
*
*
* character (&), or an end of text character (/).
*
* New-line characters have no special significance;
*
they
are
treated
as
blanks
by
the
program.
&
and
/
*
*
* should not appear in the output.
*
*
*
A
word
is
defined
as
a
nonempty
sequence
of
non-break
*
*
* characters. A break is a sequence of one or more
*
break
characters.
A
break
in
the
input
is
reduced
to
*
*
* a single blank or start of new line in the output.
*
*
*
The
input
text
is
a
single
line
entered
from
a
*
*
* simple terminal with an fixed 80 character screen
*
width.
When
the
program
is
invoked
it
waits
for
the
*
*
* user to provide input. The user types the input line, *
* followed by a / (end of text) and a carriage return. *
* The program then formats the text and displays it on *
* the terminal.
*
*
*
* If the input text contains a word that is too long to *
* fit on a single output line, an error message is
*
typed
and
the
program
terminates.
If
the
end-of-text
*
*
* character is missing, an error message is issued and *
* the user is given a chance to type in a corrected
*
version
of
the
input
line.
*
*
*
*
(end
of
specification)
*
*
********************************************************/
5.18. CHALLENGING EXERCISES
215
import java.io.*;
public class Buggy
{
final static int LINESIZE = 31;
public static void main(String [] args)
{
int k,
bufpos,
fill,
maxpos = LINESIZE;
char cw,
blank = ’ ’,
linefeed = ’$’,
eotext = ’/’;
boolean moreinput = true;
char [] buffer = new char [LINESIZE];
bufpos = 0;
fill = 0;
while (moreinput)
{
cw = gchar();
if (cw == blank || cw == linefeed || cw == eotext)
{
if (cw == eotext) moreinput = false;
if ((fill + 1 + bufpos) <= maxpos)
{
pchar(blank);
fill = fill + 1;
}
else
{
pchar(linefeed);
fill = 0;
}
for (k = 0; k < bufpos; k++) pchar(buffer[k]);
fill = fill + bufpos;
bufpos = 0;
}
else if (bufpos == maxpos)
{
moreinput = false;
System.out.println("Word to long");
}
216
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
else
{
bufpos = bufpos + 1;
buffer[bufpos-1] = cw;
}
}
pchar(linefeed);
return;
}
// I use B as a shorthand for the character ’ ’.
final static char B = ’ ’;
final static int ILENGTH = 80;
// Make suitable array with initial contents Z
// then a load of blanks.
static char [] buffer = {
’Z’,B,B,B,B,B,B,B,B,B,
B,B,B,B,B,B,B,B,B,B,
B,B,B,B,B,B,B,B,B,B,
B,B,B,B,B,B,B,B,B,B,
B,B,B,B,B,B,B,B,B,B,
B,B,B,B,B,B,B,B,B,B,
B,B,B,B,B,B,B,B,B,B,
B,B,B,B,B,B,B,B,B,B};
// bcount is defined here so that it keeps its values
// across several calls to gchar().
static int bcount = 1;
static char gchar()
{
char [] inbuf = new char [ILENGTH];
char eotext = ’/’;
char c;
if (buffer[0] == ’Z’)
{
getrecord(inbuf);
5.18. CHALLENGING EXERCISES
217
// indexOf returns the index of a position where the given
// character is present in a string, or -1 if it is not
// found.
if (new String(inbuf).indexOf((int) eotext) == -1)
{
System.out.println("No end of text mark");
buffer[1] = eotext;
}
else for (int j=0; j<ILENGTH; j++)
buffer[j] = inbuf[j];
}
c = buffer[bcount-1];
bcount = bcount + 1;
return c;
}
// a static ouput buffer, again blank-filled.
static char [] outline =
{ B,B,B,B,B,B,B,B,B,B,B,B,B,B,B,B,B,B,B,B,
B,B,B,B,B,B,B,B,B,B,B
};
// i indicates which place in outline pchar should
// put the next character at.
static int i = 1;
static void pchar(char c)
{
int linefeed = ’$’;
if (c == linefeed)
{
System.out.println(outline);
for (int j=0; j<LINESIZE; j++)
outline[j] = B;
i = 1;
}
else
{
outline[i-1] = c;
i = i + 1;
}
}
218
CHAPTER 5. DESIGNING AND TESTING PROGRAMS IN JAVA
// Get access to keyboard input. No tricks here!
static BufferedReader in =
new BufferedReader(
new InputStreamReader(System.in),
1);
static void getrecord(char [] b)
{
String s;
try
{
s = in.readLine();
}
catch (IOException e)
{
s = " ";
}
for (int i = 0; i < ILENGTH; i++)
{
if (i < s.length()) b[i] = s.charAt(i);
else b[i] = ’ ’;
}
}
}
/* End of file */
Chapter 6
A representative application
The purpose of this chapter is not to form part of the official examinable course,
but to provide you with some extended samples of Java so that you can see the
various facilities working together and so you can consider how the code ought
to have been written to make it as clear and robust as possible. Note that I do
not guarantee that my code is a paragon of clarity, and although the example
here is much larger than the ones that have gone before it has still been trimmed
fairly close to the bone to make it as small and perhaps comprehensible as I could
manage.
6.1 A Lisp interpreter
This final example for these notes is as large and complicated as any of the others
here — but most of the Java features it uses are ones that have been seen before.
It is an implementation of a very much cut-down version of the programming language Lisp. In this language, which is one of the oldest programming languages
that is still in use today, and a version of which is used to customise the emacs editor, all syntax is indicated with explicit parentheses. The programs that one writes
are very much like ML ones, except that Lisp does not have either the syntax and
operators of ML nor the type-checking. While I do not want to divert this Java
course into one on Lisp I will give a brief example of the sort of program that can
be used to test my Minilisp. It defines a function to reverse lists and demonstrates
its use.
(defun reverse (x)
(rev1 x nil))
(defun rev1 (a b)
(cond
219
CHAPTER 6. A REPRESENTATIVE APPLICATION
220
((eq a nil) b)
(t (rev1 (cdr a) (cons (car a) b)))))
(reverse ’(a b c d))
To find out about Lisp in its modern and very large form you could check
out Common Lisp — the Language[21] however the Minilisp here perhaps makes
better sense with reference to the Lisp 1.5 manual[13], which is amazingly ancient
now but which has the huge advantage of brevity. Of course the Common Lisp
manual was written by Guy Steele who participated in writing the Java language
reference — so as well as Lisp being a very direct ancestor of ML it can also be
seen as having had noticable input into Java1 . Another reason for including an
implementation of it here.
I have in fact extended the tiny Lisp shown here into a full-scale one that is
capable of running programs that are many tens of thousands of lines long. The
write-up of that work is in “Further evaluation of Java for Symbolic Calculation”,
Proc. ISSAC 00, St Andrews, Scotland, August 2000.
1 The
“Flavours” package developed for Lisp at MIT was one of the earlier programming systems that encouraged object oriented design, supported inheritance and worked to make large-scale
programming practical. CLOS (Common Lisp Object System) is the major internationally standardised model for dynamic object-oriented programming.
6.1. A LISP INTERPRETER
221
Now for the implementation. Making sense of it is liable to be a substantial struggle for most of you, but I hope that the fact that I can fit an implementation of a
programming language into these notes is at least interesting!
//
//
//
//
//
//
//
//
//
//
//
//
//
Minilisp
Basic and utterly tiny Lisp system coded
in Java by Arthur Norman, 1998.
In spirit much like an older BCPL then C
version of the same thing!
Supports
quote, cond, defun
atom, eq, car, cdr, cons
has fragments of code waiting to be extended
to do rather more.
import java.io.*;
//
//
//
//
Lisp has a single inclusive data-type, which I call
LispObject here. It has sub-types that are symbols,
numbers, strings and lists. Here I give just two
methods (print and eval) that may be used on anything.
abstract class LispObject
{
public abstract void print();
public abstract LispObject eval(Environment env);
}
// A "cons" is an ordered pair. In ML terms it would be
// a bit like (’a * ’b)
class Cons extends LispObject
{
// The left and right parts of a pair are called
//
CAR and CDR
public LispObject car, cdr;
Cons(LispObject car, LispObject cdr)
{ this.car = car; this.cdr = cdr; }
222
CHAPTER 6. A REPRESENTATIVE APPLICATION
// Function calls are written as lists (fn a1 a2 ...)
public LispObject eval(Environment env)
{
int n = 0;
for (LispObject a=cdr;
a instanceof Cons;
a = ((Cons)a).cdr) n++;
LispObject [] args = new LispObject [n];
n = 0;
for (LispObject a=cdr;
a instanceof Cons;
a = ((Cons)a).cdr) args[n++] = ((Cons)a).car;
// Now I have unpicked the actual arguments into a vector
if (car instanceof Symbol)
{
Symbol f = (Symbol)car;
// "special" functions are for QUOTE, CONS and DEFUN. They
// do not evaluate their arguments
if (f.special != null)
return f.special.op(args, env);
// All other functions have their arguments evaluated.
for (int i=0; i<n; i++)
args[i] = args[i].eval(env);
// Call the function!
return f.fn.op(args);
}
// return NIL if I do not otherwise know what to do
else return Minilisp.nil;
}
// Lists print as (a b c ... )
// and if a list ends in NIL then it is displayed with
// just a ")" at the end, otherwise the final atom is
// shown after a "."
public void print()
{
LispObject x = this;
String delim = "(";
while (x instanceof Cons)
{
System.out.print(delim);
delim = " ";
((Cons)x).car.print();
x = ((Cons)x).cdr;
}
6.1. A LISP INTERPRETER
223
if (x != Minilisp.nil)
{
System.out.print(" . ");
x.print();
}
System.out.print(")");
}
}
// I do not do a lot with strings here.
class LispString extends LispObject
{
public String string;
LispString(String s)
{ this.string = s; }
public LispObject eval(Environment env)
{ return this; }
public void print()
{
System.out.print("\"" + string + "\"");
}
}
class Symbol extends LispObject
{
public String pname;
// print name
public LispObject plist; // property list (unused)
Symbol obListNext;
// chaining of symbols
public LispFunction fn; // function (if any)
public SpecialFunction special; // special fn (if any)
// intern() looks up a Java String and find the Lisp
// symbol with that name. It creates it if needbe.
public static Symbol intern(String name,
LispFunction fn, SpecialFunction special)
{
Symbol p;
for (p=Minilisp.obList; p!=null; p=p.obListNext)
{
if (p.pname.equals(name)) return p;
}
224
CHAPTER 6. A REPRESENTATIVE APPLICATION
// not found on "object-list" (oblist), so create it
p = new Symbol();
p.pname = name;
p.plist = Minilisp.nil;
p.obListNext = Minilisp.obList;
Minilisp.obList = p;
p.fn = fn != null ? fn : new Undefined(name);
p.special = special;
return p;
}
// The symbols NIL and T are special - they evaluate
// to themselves. All others get looked up in an
// environment that stores current values of local vars.
public LispObject eval(Environment env)
{
if (this == Minilisp.nil ||
this == Minilisp.lispTrue) return this;
return env.eval(this);
}
public void print()
{
System.out.print(pname);
}
}
// An environment is a chain of Bindings terminated with
// a NullEnvironment. Each binding holds information of
// the form
//
variable = value
abstract class Environment
{
public abstract LispObject eval(Symbol name);
}
class NullEnvironment extends Environment
{
public LispObject eval(Symbol name)
{
System.out.println("Undefined variable: " +
name.pname);
System.exit(1);
return null;
}
}
6.1. A LISP INTERPRETER
225
class Binding extends Environment
{
public Symbol name;
public LispObject value;
public Environment next;
Binding(Symbol name, LispObject val, Environment next)
{
this.name = name;
this.value = val;
this.next = next;
}
public LispObject eval(Symbol x)
{
if (x == name) return value;
else return next.eval(x);
}
}
// I do not do a lot with numbers here!
class LispNumber extends LispObject
{
public int value;
LispNumber(int value)
{ this.value = value; }
public LispObject eval(Environment env)
{ return this; }
public void print()
{
System.out.print(value);
}
}
// Each built-in function is created wrapped in a class
// that is derived from LispFunction.
abstract class LispFunction
{
public abstract LispObject op(LispObject [] args);
}
226
CHAPTER 6. A REPRESENTATIVE APPLICATION
class Undefined extends LispFunction
{
String name;
Undefined(String name)
{ this.name = name; }
public LispObject op(LispObject [] args)
{
System.out.println("Undefined function " + name);
System.exit(1); // throw?
return null;
}
}
// If a symbol has an interpreted definition its
// associated function is this job, which knows how to
// extract the saved definition and activate it.
class Interpreted extends LispFunction
{
LispObject a, b;
Environment env;
Interpreted(LispObject a,
// formal args
LispObject b,
// body
Environment env) // environment
{
this.a = a;
this.b = b;
this.env = env;
}
public LispObject op(LispObject [] args)
{
LispObject a1 = a;
int i = 0;
Environment e = env;
while (a1 instanceof Cons)
{
e = new Binding(
(Symbol)((Cons)a1).car, args[i++], e);
a1 = ((Cons)a1).cdr;
}
return b.eval(e);
}
}
6.1. A LISP INTERPRETER
227
// Similar stuff, but for "special functions"
abstract class SpecialFunction
{
public abstract LispObject op(LispObject [] args,
Environment env);
}
// (quote xx) evaluates to just xx
class QuoteSpecial extends SpecialFunction
{
public LispObject op(LispObject [] args,
Environment env)
{
return args[0];
}
}
// (cond (p1 e1)
//
(p2 e2)
//
(p3 e3) )
//
if p1 then e1
else if p2 then e2
else if p3 then e3
else nil
class CondSpecial extends SpecialFunction
{
public LispObject op(LispObject [] args,
Environment env)
{
for (int i=0; i<args.length; i++)
{
Cons x = (Cons)args[i];
LispObject predicate = x.car;
LispObject consequent = ((Cons)x.cdr).car;
if (predicate.eval(env) != Minilisp.nil)
return consequent.eval(env);
}
return Minilisp.nil;
}
}
228
CHAPTER 6. A REPRESENTATIVE APPLICATION
// (defun name (a1 a2 a3) body-of-function)
class DefunSpecial extends SpecialFunction
{
public LispObject op(LispObject [] args,
Environment env)
{
Symbol name = (Symbol)args[0];
LispObject vars = args[1];
LispObject body = args[2];
name.fn = new Interpreted(vars, body, env);
return name;
}
}
// like ML
"fun car (a :: b) = a;"
class CarFn extends LispFunction
{
public LispObject op(LispObject [] args)
{
return ((Cons)(args[0])).car;
}
}
// like ML
"fun cdr (a :: b) = b;"
class CdrFn extends LispFunction
{
public LispObject op(LispObject [] args)
{
return ((Cons)(args[0])).cdr;
}
}
// like ML
"fun atom (a :: b) = false | atom x = true;"
class AtomFn extends LispFunction
{
public LispObject op(LispObject [] args)
{
return args[0] instanceof Cons ? Minilisp.nil :
Minilisp.lispTrue;;
}
}
6.1. A LISP INTERPRETER
// (eq a b) is true if a and b are the same thing
class EqFn extends LispFunction
{
public LispObject op(LispObject [] args)
{
return args[0]==args[1] ? Minilisp.lispTrue :
Minilisp.nil;
}
}
// like ML
"fun cons a b = a :: b;"
class ConsFn extends LispFunction
{
public LispObject op(LispObject [] args)
{
return new Cons(args[0], args[1]);
}
}
// (stop) exist from this Lisp.
class StopFn extends LispFunction
{
public LispObject op(LispObject [] args)
{
System.exit(0);
return null;
}
}
// The top-level class has a bunch of input
// and management code.
public class Minilisp
{
public static Symbol nil, lispTrue,
obList, lambda, cond, quote, defun;
static StreamTokenizer input;
static int inputType;
static boolean inputValid;
229
230
CHAPTER 6. A REPRESENTATIVE APPLICATION
static void initInput()
{
input = // Get stream & establish syntax
new StreamTokenizer(
new BufferedReader(
new InputStreamReader(System.in),
1));
input.eolIsSignificant(false);
input.ordinaryChar(’/’);
input.commentChar(’;’);
input.ordinaryChar(’\’’);
input.quoteChar(’\"’);
input.ordinaryChar(’.’); // disable floating point
input.lowerCaseMode(true);
inputValid = false;
}
// read a single parenthesised expression.
// Supports ’xx as a short-hand for (quote xx)
// which is what most Lisps do.
// Formal syntax:
//
read => SYMBOL | NUMBER | STRING
//
=> ’ read
//
=> ( tail
//
tail => )
//
=> . read )
//
=> read readtail
static LispObject read() throws IOException
{
LispObject r;
if (!inputValid)
{
inputType = input.nextToken();
inputValid = true;
}
switch (inputType)
{
case StreamTokenizer.TT_EOF:
throw new IOException("End of file");
case StreamTokenizer.TT_WORD:
r = Symbol.intern(input.sval, null, null);
inputValid = false;
return r;
6.1. A LISP INTERPRETER
231
case StreamTokenizer.TT_NUMBER:
r = new LispNumber((int)input.nval);
inputValid = false;
return r;
case ’\"’: // String
r = new LispString(input.sval);
inputValid = false;
return r;
case ’\’’:
inputValid = false;
r = read();
return new Cons(quote, new Cons(r, nil));
case ’(’:
inputValid = false;
return readTail();
case ’)’:
case ’.’:
inputValid = false;
return nil;
default:
r = Symbol.intern(
String.valueOf((char)inputType), null, null);
inputValid = false;
return r;
}
}
static LispObject readTail() throws IOException
{
LispObject r;
if (!inputValid)
{
inputType = input.nextToken();
inputValid = true;
}
switch (inputType)
{
case ’.’:
inputValid = false;
r = read();
if (!inputValid)
{
inputType = input.nextToken();
inputValid = true;
}
232
CHAPTER 6. A REPRESENTATIVE APPLICATION
if (inputType == ’)’) inputValid = false;
return r;
case StreamTokenizer.TT_EOF:
throw new IOException("End of file");
case ’)’:
inputValid = false;
return nil;
default:r = read();
return new Cons(r, readTail());
}
}
// set up fixed definitions
static void initSymbols()
{
obList = null;
nil = Symbol.intern("nil", null, null);
nil.plist = nil;
Symbol.intern("car", new CarFn(), null);
Symbol.intern("cdr", new CdrFn(), null);
Symbol.intern("cons", new ConsFn(), null);
Symbol.intern("atom", new AtomFn(), null);
Symbol.intern("eq", new EqFn(), null);
Symbol.intern("stop", new StopFn(), null);
lispTrue = Symbol.intern("t", null, null);
// lambda is ready for extension of this code
lambda = Symbol.intern("lambda", null, null);
cond = Symbol.intern("cond", null,
new CondSpecial());
quote = Symbol.intern("quote", null,
new QuoteSpecial());
defun = Symbol.intern("defun", null,
new DefunSpecial());
}
public static void main(String [] args)
{
initInput();
initSymbols();
System.out.println("Arthur’s Minilisp...");
6.1. A LISP INTERPRETER
233
try
{
// this is s READ-EVAL-PRINT loop
for (int i=1;;i++)
{
System.out.print(i + ": ");
// Ensure that the prompt gets displayed.
System.out.flush();
LispObject r = read();
LispObject v = r.eval(new NullEnvironment());
System.out.print("Value: ");
v.print();
System.out.println("");
}
}
catch (IOException e)
{
System.out.println("IO exception");
}
System.out.println("End of Lisp run. Thank you");
}
}
// End of Minilisp.java
6.1.1 Exercises
ML to Lisp
Find a Lisp 1.5 manual and/or study the Minilisp code, and then see how many
ML list-processing functions you can convert into Lisp and run on the Java implementation. You have already seen reverse, so the next thing to try is append. It
would probably be possible to coax the ML exercise on transitive closures through
Minilisp!
Add a few more functions
Show that you have understood what is going on in the Minilisp code by adding in
support for arithmetic, specifically functions to add, subtract, multiply and divide
numbers, and to compare them for inequality and “less-than”.
234
CHAPTER 6. A REPRESENTATIVE APPLICATION
Emacs Lisp
Now you have at least minimal exposure to Lisp, investigate the way it is used
in the emacs editor to allow users to create new language-specific editing modes,
indentation and colour conventions.
WeirdX
Visit the web-site http://www.jcraft.com/weirdx/ and fetch yourself a
copy of the source of the WeirdX X-windows server. It is around 25000 lines
of Java! Do not fetch or examine in any way the related but commercial product
called WiredX. Inspect the associated GNU public license carefully: it gives you
permission to work on the source but imposes an obligation to make the source
version of any adjustments available to the world at no cost. If you are happy with
the GNU rules2 investigate the behaviour of WeirdX and look for ways to
1. Identify and remove bugs;
2. Make it implement the latest X-windows specification more fully;
3. Enhance its performance;
4. Put a proper and copious number of comments into the code (!);
5. Ensure that it implements all possible facilities to reduce the security vulnerabilities that X-windows often opens up;
6. Make a careful comparison between the behaviour and capabilities of the
improved WeirdX and other commercial and free X servers. Create a nicely
structured wish-list for future enhancements.
As you make and test changes arrange to make your improvements available to
the the world: see http://sourceforge.net/projects/weirdx/. In case
you had not spotted, this is not a small exercise for an individual to attack lightly:
it is a substantial challenge that calls for a lot of study way beyond core Java and
if you decide to try it I suggest you form a small group to work in. Please let me
know of any progress.
2 For
an extended discussion of the GNU public license and “free” software in general, see
“The Cathedral and the Bazaar” by Eric Raymond[20]. Some of the explanation there makes very
good sense, some seems to me to be silly!
Chapter 7
What you do NOT know yet
These days anybody arranging a lecture course is expected to think in terms of
aims, objectives and learning outcomes. In particular it is deemed important to
consider carefully what will be known by students who have taken the course. I
want to take a contrary view: that what is most important to understand is what
you will not know just because you have attended this course and done all the
exercises. I view a recognition of ones limitations as vital both for any individual’s
personal integrity and as something that is essential if they are to be productive in
any work environment. So let me start with:
After one course you are not a Java expert. . .
To a large extent you only become an expert in any sort of programming after
you have built up a substantial body of experience working on both individual and
group projects. Most people can only gain the paranoia about bugs and documentation that is really needed via personal involvement in projects that fail horribly!
Most people only gain a proper paranoid attitude to overall system design via personal involvement in projects that collapse through being inadequately specified,
ill-planned or where project management is not strong enough. The Computer Science Tripos provides courses on Software Engineering that document examples of
software disasters and explain the state of the art in avoiding trouble. Despite this
most people view the horror stories as things that happen to “other people” until
it is too late.
Java is an object oriented language. Proper use of it involves obtaining full
leverage from the package, class and inheritance features it provides. This course
has taken the attitude that Object Oriented Design is something that can not be
fully appreciated until you are able to write a range of small programs comfortably. So once you have mastered this material it will be proper to re-start Java
from the beginning concentrating on starting the design by planning a class hierarchy. This has to include careful thought about private, protected etc class
235
236
CHAPTER 7. WHAT YOU DO NOT KNOW YET
members, and it can involve formal schemes to document structures.
I have explained what packages are, but not discussed the practicalities of
using lots of different packages for your own code. For small programs this is
OK, but larger scale work will put more pressure on this side of things.
The Java libraries that support windows have been introduced in a very sketchy
way here. That is because there is a huge amount to understand, and it would not
even start to fit in the time available within the Part IA course. All the issues
of getting pages laid out, controlling multiple windows, organising cut-and-paste
operations and so on take a lot of learning. The most that can be hoped is that this
course has given you a starting point to work from.
When a Java applet is run using a web browser it is subject to a variety of
security limitations. This is so that when some remote user loads a web page
that runs your Java code you can not then steal or destroy information on their
computer. There is an elaborate scheme involving signatures and permissions that
makes it possible for applets to be granted additional privilege, without giving
them total access. Such issues will be important for large-scale Java projects.
Network access and concurrency are both areas where a programmer needs
to absorb quite a lot of additional understanding before they can use Java (or any
other language) in a fully satisfactory manner. It is easy to end up with systems
that can either gum up in deadlock or have different threads create inconsistent
results, and proper recovery after one thread fails or one network link times out
is not at all easy to arrange. The operating systems threads later in the CST have
much to say about the issues and pitfalls involved. The way in which Java can
interface to databases may appear straightforward, but competent design of the
database itself needs specialist understanding.
While Java is a pretty respectable general purpose language there are plenty of
areas where simple use of pure Java will prove inadequate. Sometimes however
it will still make sense to implement either the bulk of a program or perhaps just
the user-interface in Java, with some other components in another language. This
raises issues of inter-language working, and serious design challenges in the area
of the inter-language interface.
During this course you have been encouraged to use javac as your Java compiler. For real projects it is almost certain that much more elaborate tools would
be used. These would include ones to manage a chain of versions of code, tools to
generate stylised code for some bits of a user interface, and a variety of debugging
and performance analysis packages.
Actually hardly any programmer, however experienced, will be a full expert
on all of the above matters. . .
. . . but at least you have taken the first step!
Chapter 8
Model Examination Questions
Some of these are ones I have invented, some have been submitted by members
of the class, while yet more are adjustments of previous examination questions
modified to fit into a Java context. They may not all be normalised for precision
of wording or difficulty, but should give you something to try your skills on. I
neither confirm nor deny the possibility that variants on some of these may appear
in this or a future year’s real examination paper. . .
Note also that for the Computer Science Part IA papers there can be full-sized
(ie around 30 mins) questions, half size (around 15 mins) and tiny (about 1 minute)
questions, and some of these fit or could readily be modified to fit several of these
categories.
8.1 Java vs ML
Compare and contrast Java with ML under the following headings
1. Primitive data types;
2. Support for arrays;
3. Support for lists and trees;
4. Functions applicable to several different data-types;
5. Complexity of syntax.
[4 marks per section]
237
238
CHAPTER 8. MODEL EXAMINATION QUESTIONS
8.2 Matrix Class
Design a Matrix class for doing operations on n × n matrices.
The class should include functions for performing matrix multiplication, addition, and multiplication by scalars (and all matrices may be supposed to have
elements that are type double).
To what extent should other classes have rights to modify individual matrix
elements? Justify your answer.
8.3 Hash Tables
Java allows you to declare arrays with any sort of content — for instance int,
double, String or Object, but the value used to index the array is restricted
to being an int. In some applications programmers want structures that behave
like arrays whose subscripts are of type String. One way of doing this is to
use a structure known as a hash table: Given an index value of type String an
integer is computed using a hash function. The Java method hashCode in class
String computes a suitable value. The int value computed depends only on the
contents of the String, but different strings can be expected to lead to integers
that are uniformly distributed across the whole range that integers cover. This
hash value is reduced modulo the size of a fixed ordinary array, this (almost) lets
the original String act as an index. The problem that arises is that it could be
that two different String values hash onto the same location in the array.
This concern can be overcome by making the entries in the array linked lists
of (key, value) pairs, so that the original index String will match one of the
key strings and then the associated value is what is stored with it. The method
equals(String s) in class String returns a boolean telling if one string is
equal to another. Storing into a hash table will involve adding a new (key, value)
pair to one of the lists.
Design appropriate data structures and classes for such a table in the case
where the values to be stored are themselves of type String. Use the following
method signatures in a class called HashTable:
void put(String key, String value) throws Duplicate;
String get(String key) throws Missing;
where you may assume that the exceptions have already been defined elsewhere.
Note: The Java class-libraries provide a class called HashMap that does all this
for you! The exercise here in coding it for yourself not something the ordinary
Java programmer ever needs to do!
8.4. COMPASS ROSE
239
8.4 Compass Rose
A Java program to draw compass
roses is needed. In the example
shown the line pointing North is 6
units long, that pointing South is 5,
East and West 4, NE, SE, SW and
NW 3 and NNE, ENE and so on 2
and the rest 1 unit. In general if the
line pointing North is n units long
there will be 2n−1 radial lines in all in
the complete rose. You may assume
that in your Java applet the method
paint will be called when an image
is to be displayed, and it gets passed
an argument of type Graphics. The
class Graphics has a method called
drawLine that draws a line from
one point to another given four integer arguments x1 , y1 , x2 and y2 .
Write two versions of the compass rose applet. The first should use
recursion on the length n of a line,
while the second should be iterative.
In the second case it may be useful
to write an auxiliary method that calculates the highest power of two that divides
exactly into a given number.
The length of the North line should be specified as a final variable in the
class.
8.5 Language Words
In the context of Java, and given that this whole question is supposed to last 30
minutes, explain the following:
• package
• class
• import
• public
CHAPTER 8. MODEL EXAMINATION QUESTIONS
240
• protected
• interface
• static
8.6 Exception abuse
Show how Java exceptions can be used in conjunction with a for loop
1. to simulate the effect of a break; statement;
2. to simulate the effect of a continue; statement.
8.7 Queues
A priority queue maintains a list of pairs, each consisting of a priority (an integer)
and a name (a String), sorted into increasing order of priority. Three operations
are required — one that just creates an empty queue, and then one called insert
that takes an integer and a string as its arguments and places them in the correct
position in the list. Finally a method next takes no arguments. It removes the
first pair from the list and returns its text string. It should throw an exception if
the queue is empty when it is called.
Give the definition of a Java class that implements the queue object.
8.8 Loops
Describe the features of Java for controlling the repeated execution of a block of
code.
Show how general uses of for, while and do could all be emulated using
only loops that start of while (true).
8.9 Snap
Two identical packs of ordinary playing cards (52 different cards per pack) are
shuffled and places face downwards on a table. Two players then play a game
of Snap. Each is allocated one pack, and in each turn in the game one card is
turned up. The two upturned cards (one in front of each player) are compared.
If the cards match a snap-turn is declared. A game ends when all 52 cards have
been compared. Being computer scientists rather than five year olds these players
8.10. PARTITIONS
241
just record snap-turns and do not pick up or otherwise disturb the cards when one
occurs!
Write a Java program which will simulate the game for the purposes of determining the probability of there being at least one snap-turn in a game. You may
assume the existence of a random number generator but must state the properties
of it that you rely on.
8.10 Partitions
Write a program in Java which, given two integer inputs j and k will output the
combinations of k things partitions into k groups. For instance if j = 5 and k = 3
the output would be
(5,
(4,
(3,
(3,
(2,
0,
1,
2,
1,
2,
0)
0)
0)
1)
1)
8.11 Laziness
A Java class as follows has been defined
abstract class NextFunction
{
public int next(int n);
}
class Lazy
{
int head;
Lazy tailOrNull;
NextFunction next;
public Lazy(int head, NextFunction next)
{
this.head = head;
this.tailOrNull = null;
this.next = next;
}
public int first() { return head; }
public Lazy tail()
{
if (tailOrNull != null) return tailOrNull;
tailOrNull = new Lazy(next(head), next);
CHAPTER 8. MODEL EXAMINATION QUESTIONS
242
return tailOrNull;
}
}
The idea is to use this to represent lazy lists. In fact the small trick in the
tail function that checks if the successor to a node has already been computed
makes this a good representation. Derive a sub-class from NextFunction that
overrides the next method with one that allows you to create a lazy list of integers (1,2,3,...). Write code that, when given a lazy list, will print the first n
integers in it.
Adapt the code so as to create a function that can accept a lazy list and generate
a new lazy list from it that holds values which are the squares of the ones in
the original list. Thus if passed my first lazy list as an argument this one would
generate (1,4,9,...).
8.12 Cryptarithmetic
Write a Java program which can solve cryptarithmetic puzzles in the format of the
sum of two words. For example given the input
SEND
+MORE
----MONEY
the program would output
9567
+1085
----10652
NB Each letter has to represent a different digit.
8.13 Bandits
In Snoresville in small-town America, there are N bandits, and only one sheriff, Sheriff Dozy, who likes to do the minimum amount of work possible. He
knows quite a lot about each of them, to the extent that he knows precisely which
bandits each bandit knows. Soon, $200,000 worth of gold will be deposited in
Snoresville’s main bank; a very tempting target for the local thieves. Dozy knows
that the defences and procedures are more than good enough to resist attack by
8.13. BANDITS
243
one bandit, and that they can even resist attack by N/3 bandits. He wonders if
some sub-set of the collection of local bandits will manage to gang together to
grab the gold.
The friendship data for the bandits can be represented in the following way: A ‘1’ indicates a friendship with another bandit. If one bandit is known 1
N = 5
then he also knows of the bandit who knows him, i.e. 01
friendship is mutual. Each line ends with a ‘1’ because 111
each bandit knows himself, obviously. Explain why 0101
10011
the triangular table given is all that’s needed to detail
all the friendships, and show how it can be expanded
into a square to give the friendship data with a row for
each bandit. [2 marks]
Given a global data structure defined as an N by N matrix (int [][]gang),
filled initially with zeros, and where N is the largest number of bandits to consider,
give two Java procedures:
1. generate(): This should input from the keyboard an integer n, and a
floating-point p. Check that n is in the range 0 < n ≤ N, and that p is a valid
probability between 0 and 1. The procedure should then randomly generate
a triangular table as above, such that the probability of any of the points
being ‘1’ is p, except for each bandit with himself which should always be
‘1’.
2. square(): This should take the triangular table now in gang from the
generate function and transform it into a square table as discussed earlier. It
should then display the block in a tabular form. [6 marks]
3. A group of bandits is only possible when each bandit knows every other
bandit in the group. In the example given above, what is the group with
largest cardinality? [1 mark]
4. Examine the following pieces of code; the procedure compare and a fragment of code to show how it is called.
Calling Fragment:
for (i=0;i<n;i++)
{
compare(i,gang[i]);
for(j=0;j<n;j++)
{
gang[i][j]=0;
gang[j][i]=0;
}
CHAPTER 8. MODEL EXAMINATION QUESTIONS
244
}
Main example
int maxgroup[N];
int max=0;
/*
* NOTE: current[N] is one array of the array of
* arrays gang[N][N]
*/
void compare(int i, int current[N])
{
for (int j=0; j<n; j++) current[j] &= gang[i][j];
for (int k=i+1; k<n; k++)
{
if (current[k]==1)
{
compare(k,current);
current[k]=0;
}
}
if (count(current)>max)
{
max=0;
for (int k=0; k<n; k++)
{
if (current[k]==1) maxgroup[max++]=k;
}
}
}
5. Explain in detail what the function compare does, and how it works, paying
special attention to the recursive nature of the procedure and the actions of
the calling fragment, and giving a function count which returns the number
of ‘1’s in the array current. [8 marks]
6. Outline a function main to bring all this together and for each N in the range
2 to 20 calculate an average, over 10 tries, of the cardinality, given p = 0.5,
and display the smallest N for which the cardinality is less than N/3. This
is the critical number of bandits for the gold safety, and for sheriff Dozy not
to lose his job. [3 marks]
8.14 Exception
Describe some circumstances where it is useful for functions to return errors as
exception, and some where it is not. Give an example of an algorithm which is
simplified by the use of exceptions.
8.15. FEATURES
245
8.15 Features
Write brief notes on four of the following aspects of Java. In some cases it may be
appropriate to compare what Java does in the situation with other programming
languages such as ML.
1. Using the same name for several different functions;
2. Data types where a single type has several variants;
3. Programs that live in several source files;
4. Inheritance and abstract classes;
5. The degree to which code will behave the same when run on different computers.
8.16 More features
For five of the following Java features write a very short code fragment (it does not
have to be complete, and perhaps 2 or 3 lines will suffice in most of the cases) that
illustrates the syntax involved. In each case explain briefly what your example
achieves.
1. Declaration of constants;
2. Casts between class types;
3. The two styles of comment;
4. Catching an exception;
5. The switch statement, including a default label;
6. Summing all the BigInteger values that are in an array;
7. The class Object.
CHAPTER 8. MODEL EXAMINATION QUESTIONS
246
8.17 Debate
A grand debate is being planned by a society that has among its members a large
number of computer professionals and working programmers. Arthur will propose a motion “That this house considers ML to be a much better programming
language than Java”, while Larry will lead the opposition under the banner “Java
is the language for any programmer who seeks employment and hence who can
truly be referred to as working”. Organise the points in support of the two languages that each proponent will use to justify their position, and identify areas
where they are liable to find common ground.
You are not expected to reach a definite conclusion about which way the vote
will go at the end of the debate: your job is just to collect and organise the arguments so that comparing and contrasting the merits of the two language becomes
easy.
8.18 Design
You have been invited to start a project to build a program that will check ML programs to see if they have missing punctuation marks or other mistakes, so that this
check can be performed before the ML code is forwarded for full execution. So
far all you know about what is wanted is the above. Without concerning yourself
with fine detail of how the checking will be implemented, identify and discuss:
1. questions you might want to ask the client organisation about its needs before deciding on any more details of your design;
2. choices or options available to you when making a detailed project plan;
3. ways of partitioning the whole job into a handful of separate modules that
could be implemented more or less independently;
4. plans for testing the code you write and determining whether, when notionally finished, it has met its objectives.
8.19 Filter (Coffee?)
The structure of a binary tree containing integers at just some of its leaves could
have been given by the ML type T defined as follows
datatype T = X | N of int | D of T*T;
8.20. PARSE TREES
247
Define a Java class or set of classes that can be used to represent trees in the
same general form. Then add a method filter which uses an integer k and a
tree. Its job is to simplify trees by using the rule that and sub-node (type N in the
ML declaration) where the integer stored is k gets turned into an X leaf. Then any
node (X,t), or (t,X) [ie nodes of the ML type D where one of the sub-trees is
an X] get turned into just t.
Thus for instance if k = 0 the initial tree ((0,0),((2,0),3)) would simplify to just (2,3)
8.20 Parse trees
What does it mean for a Java class to be abstract?
A Java program includes the following class declarations:
abstract Class Node
{
public int eval();
}
Class Num extends Node
{
int value;
public Num(int value) { this.value = value; }
...
}
Class Op extends Node
{
String sym;
Node left;
Node right;
public Op(String sym, Node left, Node right)
{ this.sym=sym; this.left=left; this.right=right; }
...
}
The objective of the programmer who wrote this was to be able to write assignments such as
test = new Op("*", new Num(4),
new Op("-", new Num(7), new Num(2)));
The variable test is of type Node which can cover either of the two concrete
cases of Op (representing a dyadic operator together with its two operands) or Num
248
CHAPTER 8. MODEL EXAMINATION QUESTIONS
(a number). Thus the above assignment sets up a representation of the expression
4 ∗ (7 − 2).
The abstract class Node declares a method eval(). Fill in the dots in the
other two classes with code that overrides this so that calling the eval method on
a Node returns the value of the arithmetic expression is represents, supposing that
the only operators that will be used are plus, minus and times.
8.21 Big Addition
Java comes with a class BigInteger that represents potentially huge numbers.
Suppose it did not, or for some reason you were prohibited from using it but still
needed to work with large positive integers. To fit your needs you will define a
new class called Big that stores integers as arrays of byte values, where each byte
holds a single decimal digit from the number being used, with the least significant
digit held at position 0 in the array.
Write a definition of such a class including in it methods to create a big integer
from an int (provided that int is positive), to add two Big values together and
to convert from a big to a String ready for printing. You need not implement
any other methods unless they are needed by the ones mentioned here.
8.22 Lists in Java
A list in Java can be represented as a sequence of links. Each link is an object
containing one value in the list and a reference to the rest of the list following the
link. A null reference indicates the end of the list.
Write a Java class that can represent such lists, where the items stored in lists
are of type Object. Provide your implementation with two static public methods
that append lists. The first of these should be called append and should take two
arguments, its result should be the concatenation of the two lists and neither input
should be disturbed. The second should be called conc and should have the same
interface, but it should work by altering the final reference in the first list to point
it towards the second, and it should thus not need to use new at all.
8.23 Pound, Shillings and Ounces
The Imperial system for Sterling currency was based on the pound, shilling and
penny (plural pence). There were 12 pence in a shilling and 20 shillings in a
pound. A Java class that could store amounts in this format might be
8.24. DETAILS
249
class LSD
{
boolean negative;
int pounds;
int shillings;
int pence;
LSD(boolean m, int l, int s, int d)
{ ... }
...
Adjust or finish off the constructor so that it raises an exception of class
BadInput (which may be supposed to have been defined already) if the input is
invalid, ie unless the number of pence is from 0 to 11 and the number of shillings
from 0 to 19, and the specified number of pounds is positive.
Now you need to provide a toString method in the class that converts currency into textual form. The following table shows the desired effect when a
number of pence is first converted to LSD and then to a string:
0
1
10
60
80
252
479
1201
2400
-252
⇒
⇒
⇒
⇒
⇒
⇒
⇒
⇒
⇒
⇒
zero
1 penny
10 pence
5 shillings
6 shillings and 8 pence
1 pound and 1 shilling
1 pound, 19 shillings and 11 pence
5 pounds and 1 penny
10 pounds
minus 1 pounds and 1 shilling
Credit will be given for a clearly explained, concise and tidily presented solution.
Minor syntax or punctuation errors in the Java code will not count heavily against
you.
8.24 Details
Give a brief explanation of each of the following aspects of Java
1. The difference between >> and >>>;
2. The possibility that in some program the test (a == a) might return the
value false for some variable a;
3. The keywords final and finally;
250
CHAPTER 8. MODEL EXAMINATION QUESTIONS
4. The expression "three" + 3 and other expressions of a generally similar
nature;
5. The meaning of or errors in (whichever case is relevant!)
...
int [10] a;
for (int i=1; i<=10; ++i)
a[i] = 1-a[i];
...
8.25 Name visibility
A complete Java program may use the same name for several different methods or
variables. Java has a number of features that allow the user to prevent such re-use
of names from causing chaos. Describe there under the headings:
1. Scope rules within individual functions; [6]
2. Visibility of method names within classes, and the effects of inheritance; [8]
3. Avoiding ambiguity when referring to the names of classes. [6]
8.26 Several Small Tasks
Write fragments of Java definitions, declarations or code to achieve each of the
following effects. You are not expected to show the whole test of a complete
program — just the parts directly important for the task described, and you may
describe in words rather than Java syntax any supporting definitions or context that
you will want to rely on. Clarity of explanation will viewed as at least as important
as syntactic accuracy in the marking scheme. It is also understood that names of
methods from the standard Java class libraries are things that programmers check
in on-line documentation while writing code, so if you need to use any of these
you do not need to get their names or exact argument-format correct provided
that you (a) describe clearly what you are doing and (b) your use is correct at an
overview level:
1. Take a long argument called x and compute the long value obtained by
writing the 64 bits of x in the opposite order; [6]
8.27. SOME TINY QUESTIONS
251
2. Define a class that would be capable of representing simple linked lists,
where each list-node contains a string. You should show how to traverse
such lists, build them and how to reverse a list. In the case of the list reversing code please provide two versions, one of which creates the reversed list
by changing pointers in the input list, and another which leaves the original
list undamaged and allocates fresh space for the reversed version; [8]
3. Cause a line to appear in a the window of an applet running from the bottom
left of the window towards the top right. Your line should remain visible if
the user obscures and then re-displays the window, but you can assume that
the size of the windows concerned will be fixed at 100 by 100 units. [6]
8.27 Some Tiny Questions
1. List the eight Java primitive data types.
2. What result will be printed if the following fragment of Java code is executed? Why?
double d = 6.6;
try
{
d = 1.0 / 0.0;
}
finally
{
System.out.println("d = " + d);
}
252
CHAPTER 8. MODEL EXAMINATION QUESTIONS
Figure 8.1: Remember: programming is fun!
Chapter 9
Java 1.5 or 5.0 versus previous
versions
The Java course from 2005 onwards uses a version of Java (and its libraries) that
support a range of things that earlier release did not. The commentary about these
features here is not part of the examinable content of the course, but may help
those who want their code to be backwards compatible, and may help supervisors
understand how features new in 1.5 are relevant in an introductory Java course.
9.1 An enhanced for loop
New Java allows direct iteration over either arrays or collection types using syntax
along the lines of
for (Type s : arrayOrCollection) use(s)
With previous versions you would need
for (int i=0; i<array.length; i++) use(array[i])
for arrays, or the yet more clumsy
for (Iterator i=collection.iterator(); i.hasNext();)
{
Type s = (Type)i.next();
use(s);
}
There are special delicacies to watch with the use of nested iterations if you
use the old versions, and a strong interaction with the generics feature described
next.
253
254
CHAPTER 9. JAVA 1.5 OR 5.0 VERSUS PREVIOUS VERSIONS
9.2 Generics
Most of the examples that use collections in these notes decorate the type declarations of the collections to show what they contain. For instance there could
be use of HashMap<String,String> for a HashMap that will only be used
with strings as both they keys and values it stores. With these decorations code
is naturally type-safe. Older Java does not support this. The result is that instead
all generic structures use the fall-back Object type. When data is retrieved from
then the Java type-checker does not know what sort of Object is being used, and
so casts, involving run-time checks, have to be used.
9.3 assert
The assert keyword came in with Java 1.4. For use with any earlier version of
Java you must remove them all. Please ensure that your code is fully debugged
first!
9.4 Static imports
If your new code includes a line such as import java.Math.PI and you then
use the constant PI in your code you will need to remove the import statement
and use the longer name Math.PI everywhere that you reference it.
9.5 Auto-boxing
With both collections and the printf facility listed later, you can use the built
in types (eg integers and reals, characters and booleans) without having to worry
too much about special consequences of them being primitive built-in types rather
than part of the Java class system. In older versions of Java this is not the case,
and you need to make much more explicit use of the wrapper classes Integer,
Double and so on. The effect is much messier code in places! The term “autoboxing” refers to the fact that these wrapper classes are still used, and use of a
constructor new Integer(1) is referred to as “boxing” the integer up.
9.6 Enumerations
In Java 1.5 you may have used enum to introduce a collection of distinct names,
and you may then have used these enumeration values in switch statements. Be-
9.7. PRINTF
255
fore Java 1.5 you could either use an explicit encoding as integers (which does
not protect you from type errors) or some more elaborate scheme that uses the
class system to protect details of the implementation. In any case the effect is
significantly more clumsy than the new scheme.
9.7 printf
When you look at many existing Java books and sample codes you will see printing done using the idiom
System.out.println("The result is" + i);
where these notes have written something more like
System.out.printf(
"The result is %d (or %<x in hex)%n", i);
These notes have preferred the second if only because the format string item
%d makes explicit what type of item is being displayed, and so there is an extra
check on internal consistency in your code. Format conversions provide amazing
(and sometimes complicated) levels of refinement in controlling just how simple
information such as numbers are to be laid out. Replicating that control using
the facilities from previous Java releases is tedious and leads to fairly bulky and
unreadable mess.
9.8 Scanner
If you needed to split an input file into words or wanted to read in a simple column
of numbers you may have used Scanner. Beware because it is new and you
would need to do things by hand to if backwards compatibility was essential. In
big programs Scanner may not matter but I find it jolly handy in all sorts of small
examples.
9.9 Variable numbers of arguments for methods
The smallVarargs facility is something you should probably not often use directly in your own code, but it is exploited by printf and friends. The old way
of achieving a similar effect is to make your function accept an array of items, and
construct a new array to pass arguments in each call you make to it.
256
CHAPTER 9. JAVA 1.5 OR 5.0 VERSUS PREVIOUS VERSIONS
9.10 Annotations
Java 1.5 provides a general framework for adding annotations to your code in a
way that is expected to make it easy for program management tools to extract
them. This is a relative of the notation /** that inserts “document comments”
that javadoc can extract. The scheme is not mentioned in this course beyond
the fact that the new-style annotations are introduced by a name preceded by
an at-sign (@), and new styles of annotation can be declared using the keyword
@interface. Any code that has this syntax will need it removed for use with
earlier versions of Java. This will not change the behaviour of the code itself
in any way, but would have an effect on how it could interact with annotationprocessing tools (see the Java documentation for the command apt). For those
who want to investigate further it might be useful to note that the Java reflection
mechanisms can detect when a method defined in a class has been annotated. A
lot of the time you will not use annotations, but when you do they may be a huge
help and you would hate to go back to a system without them.
9.11 Enhanced concurrency control
For example ConcurrentHashMap, EnumSet.
Many of the above features interact together so that the savings of using them
combined are even greater than using them one at a time. A consequence of
that is that giving them up would be even more painful than you might at first
expect. I fully expect to see many of them becoming the standard idiom for Java
programmers in the very near-term future.
Bibliography
[1] Abramowitz and Stegun. Handbook of Mathematical Functions (with formulas, graphs and mathematical tables). Dover, 1965.
[2] David Barnes and Michael Kölling. Objects First with Java: a Practical
Introduction using BlueJ. Prentice Hall/Pearson, 2 edition, 2005.
[3] Colin Bentley. Introducing SSADM 4+. NCC Blackwell, and also see
http://www.blackwellpublishers.co.uk/ssadmfil.htm, 1996.
[4] Jon Bentley. Programming Pearls. Addison-Wesley, 1986.
[5] Jon Bentley. MoreProgramming Pearls. Addison-Wesley, 1988.
[6] Judy Bishop. Java Gently. Addison Wesley, 3 edition, 2001.
[7] Fred Brookes. The Mythical Man Month. Addison Wesley, 2 edition, 1996.
[8] J. Cameron. JSP and JSD: The Jackson Approach to Software Development.
IEEE Computer Society Press, 1989.
[9] Leiserson Cormen and Rivest. An Introduction to Algorithms. MIT and
McGraw-Hill, 1990.
[10] E. W. Dijkstra. A Discipline of Programming. Prentice-Hall, 1976.
[11] Bruce Eckel. Thinking in Java. Prentice Hall, 1998.
[12] Christopher Essex, Matt Davison, and Christian Schulsky. Numerical monsters. SIGSAM Journal, 34(4):16–32, December 2000.
[13] John McCarthy et al. The Lisp 1.5 Programmer’s manual. MIT Press, 1965.
[14] M. H. Halstead. Elements of Software Science. Elsevier North Holland,
1977.
[15] Henry S Warren Jr. The Hacker’s Delight. Addison Wesley, 2003.
257
258
BIBLIOGRAPHY
[16] Brian W. Kernighan and Dennis M. Richie. The C programming language.
Prentice-Hall, 1978.
[17] Donald E. Knuth. Literate Programming. CSLI Lecture Notes and CUP,
1992.
[18] Steve Maguire. Writing Solid Code. Microsoft Press, 1993.
[19] C.A.R. Hoare O.-J. Dahl, E.W. Dijkstra. Structured Programming. Academic Press, 1972.
[20] Eric S Raymond. The Cathedral and the Bazaar. O’ Reilly, 1999.
[21] Guy Steele. Common Lisp the Language. Digital Press, 1990.
[22] X3J11. ANSI X3.159, ISO/IEC 9899:1990. American National Standards
Institute, International Standards Organisation, 1990.
Index
3n + 1, 90
π, 47
++ and --, 71
abstract, 123
Annotations, 256
Applet, 37
Applet + application, 157
appletviewer, 42
array.length, 62
Arrays, 61
Assert, 254
assert, 88
Auto-boxing, 254
AWT, 155
BigInteger, 98, 147
Binary files, 145
Binary trees, 115
Book-list, 18
Bottom-up implementation, 189
BufferedImage, 57
BufferedReader, 118
Buggy sample code, 211
byte, short, int and long, 52
Capital letters, 65
catch, 87
char, 56
Character escapes, 56
Checking assertions, 88
close a file, 142
Code layout, 33
Collection classes, 150
Colouring by syntax, 38
Colours for drawing, 45
Command-line arguments, 61
Compiling a Java program, 34
Complex numbers, 108
Constructors, 107
Continued fractions, 139
cos, 47
crypto jurisdiction policy files, 148
Default visibility, 122
Designing and testing, 167
Dip. Comp. Sci., 16
Documentation comments, 33
Documenting, 195
Doubles as bit-patterns, 138
Dutch national flag, 67
Eight queens, 81
Emacs, 29, 34
encryption and checksums, 147
Enhanced for loop, 253
Enumerations, 254
Exam questions, 237
Extended Euclinean algorithm, 84
File access, 140
File object, 144
FileReader, 141
FileWriter, 143
final, 89, 122
finally, 87
float and double, 54
Forests of trees, 79
259
INDEX
260
Game of Life, 80
Generics, 254
getClipBounds, 112
Global Font Lock, 38
goto, 51
Nested scopes, 100
Network programming, 153
new, 63
Newline character, 56
Not a Number, 54
Hacker’s Delight, 76
Hakmem, 76
Hello World, 28, 31
Hexadecimal, 54
Hilbert matrix, 68
HTML, 42
Object, 130
Object as a class, 130
Object oriented, 125
Obtaining Java, 24
Octal, 54
Oz the Werewolf, 12
IEEE floating point standard, 54
import, 102
import static, 47
Infinities, 54
Inheritance, 115
Initialised arrays, 62
inner class, 156
InputStreamReader, 118
Integer overflow, 52
interface, 124
package, 101
Packages and jar files, 110
paint, 46
parseInt, 118
Pentominoes, 134
Pollard rho, 96
Polynomial manipulation, 93
Portability, 205
printf, 255
Printing floating point values, 56
PrintWriter, 143
private, 122
protected, 122
public, 122
public and private, 101
Puzzle with two flasks, 164
JApplet, 37
Java 1.5, 253
Java Bean, 106
JLex, 118
Legendre polynomials, 95
line numbers, 147
Lisp in Java, 219
Long integer constants, 53
Mandelbrot set, 110
Menus etc, 155
Mouse co-ordinates, 39
Mouse events, 39
MULDIV, 209
Mystery big decimal, 161
Nancy Silverton, 185
native, 124
Quicksort, 91
Random numbers, 148
RC4 encryption, 69
read characters, 141
Reflection, 130
Reserved words, 49
Rounding errors, 67
RSA encryption, 160
RSI, 194
Running a simple program, 34
Scanner, 255
INDEX
Send more money, 242
set and get methods, 105
Shift operators, 72
Sieve for primes, 77
sin, 47
Sorting, 92
Sprouts, 210
static, 122
static import, 254
StreamTokenizer, 118
String concatenation, 61
Supervisor’s guide, 16
Swing, 155
switch statements, 85
synchronized, 124
tan(x): series, 113
Tar Pits, 169
Ten recommendations, 13
Text editing, 164
this, 108
Threads, 150
throw, 88
Tickable exercise 1, 24
Tickable exercise 2, 43
Tickable exercise 3, 65
Tickable exercise 4, 90
Tickable exercise 5, 93
Tickable exercise 6, 110
Tickable exercise 7, 162
Top-down design, 186
Transpose a matrix, 209
try, 87
Turtle graphics, 46
Two’s complement, 53
Unicode, 57
Varargs, 255
Visibility qualifiers, 120
Walk-through, 193
261
Webget, 162
Weirdx, 234
Wolves and caribou, 109
Writing programs is easy, 168
Writing programs is fun, 168
Y co-ordinate, 47
zip and unzip files, 147